| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Ann Thorac Surg 2006;81:650-657
© 2006 The Society of Thoracic Surgeons
a Department of Cardiothoracic Surgery, Graduate School of Medicine, University of Tokyo, Tokyo, Japan
b Research and Development Laboratory, Nipro Co, Shiga, Japan
c Division of Cardiac Surgery, University of Western Ontario, London Health Science Centre, London, Ontario, Canada
Accepted for publication July 6, 2005.
* Address correspondence to Dr Suematsu, Department of Cardiothoracic Surgery, University of Tokyo, 7-3-1 Hongo, Bunkyo-ku, Tokyo, 113-8655, Japan (Email: suematsu{at}rogers.com).
| Dr Matsuda discloses a financial relationship with Nipro Co.
|
 |
Abstract |
|---|
 Top Abstract IntroductionMaterial and MethodsResultsCommentThe Society of Thoracic...AcknowledgmentsReferences |
|---|
METHODS: Our absorbable substitute consists of three layers: a middle layer of aterocollagen between two layers of sodium hyaluronic acid and aterocollagen. In experiment 1 in this study, the patch, made of 9,000 filaments of aterocollagen fibers, (group 1; n = 5) was compared with a patch made of 6,000 filaments (group 2; n = 7), an expanded polytetrafluoroethylene sheet (group 3; n = 6), and a control group (group 4; n = 4). Subsequently, in experiment 2, the patch was examined at 4 weeks (n = 5), 12 weeks (n = 5), and 24 weeks (n = 4) after the operation by light microscopy and scanning electron microscopy.
RESULTS: The area of adhesion in group 1 was significantly less as compared with that in the other three groups, and the coronary vessels were clearly identifiable; on the other hand, all the animals in the control group showed moderate to severe adhesions, and the coronary vessels were completely obscured. In experiment 2, formation of a membranous tissue resembling the native pericardial membrane was observed in all animals, and the thickness of this membrane showed a marked increase by 24 weeks after the operation. Light microscopy and scanning electron microscopy also showed the formation of a mesothelium-like lining.
CONCLUSIONS: The new absorbable and regenerative collagen patch seemed to be biocompatible, and its use was associated with minimal adhesion formation and preserved coronary anatomy.
Formation of retrosternal and intrapericardial adhesions after median sternotomy is a widely known phenomenon and a significant cause of morbidity and mortality [1–3]. On the other hand, resternotomy has been reported to be associated with a 2% to 6% incidence of major vascular injury [1]. In addition, adhesions formed between the pericardium and epicardium also obscure epicardial structures, such as coronary arteries and bypass grafts.
Several attempts have been made to resolve the problem of formation of pericardial adhesions after cardiac surgery by using different types of pericardial substitutes, including expanded polytetrafluoroethylene (ePTFE) patches, silicone membranes, polyurethane, and glutaraldehyde-treated bovine or porcine pericardium xenografts [4–7]. Mitchel and colleagues [8] reported that hyaluronic acid solutions effectively prevented the formation of pericardial adhesions in their experimental study. However, none of these materials has yet been routinely applicable in the clinical setting, because of the absence of clear clinical evidence of their beneficial effects.
If absorbable patches could serve as a scaffold for cell generation and be replaced with native tissue, reconstruction with these materials may be an attractive alternative to the use of ePTFE patchs and xenografts. Autologous pericardium can also be used for reconstruction or repair of cardiac anomalies.
We considered that the new absorbable patch made up of collagen and hyaluronic acid may allow pericardial regeneration with minimal adhesion formation and preservation of the coronary anatomy. Therefore, the purpose of this study was (1) to evaluate the efficacy of this material with that of the ePTFE patch for minimizing postoperative adhesion formation in comparison with that in a control group in which pericardium was left open, and (2) to evaluate the course of tissue regeneration after implantation of the new absorbable patch.
|
Material and Methods |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentThe Society of Thoracic...AcknowledgmentsReferences |
|---|
New Absorbable Patch
The new absorbable patch consists of three layers, and is made entirely of a pepsin extract of the aterocollagen of swine skin with sodium hyaluronic acid. The middle layer, which serves as a scaffold for pericardial regeneration, is sheetlike and is designed to resolve within a month. The other two layers, which serve as antiadhesion materials, are designed to resolve within a week. Cross-linking of the aterocollagen was induced by thermal dehydration, and the sheet was sterilized by gamma rays. We prepared two different patches with different counts of the collagen filaments. A collagen filament was made by wet spinning, and the unwoven cloth as the middle layer was made using 9,000 or 6,000 collagen filaments in this study. The patch used in group 1 had 9,000 filaments of aterocollagen fibers in the middle layer, and that used for group 2 had 6,000 fibers of aterocollagen in the middle layer. The finished product was provided by Nipro (Shiga, Japan).
Surgical Protocol
A total of 22 beagles with a median weight of 11.0 kg (range, 10.0 kg to 12.8 kg) were used for experiment 1 of this study. General anesthesia was induced by intramuscular injection of ketamine hydrochloride (30 mg/kg) and intravenous injection of sodium pentobarbital (20 mg/kg), followed immediately by endotracheal intubation; anesthesia was maintained by additional intravenous injections of sodium pentobarbital (5 mg/kg) and 2% isoflurane. Respiration was maintained by artificial ventilation in the volume-control mode.
After standard skin preparation, a left lateral thoracotomy was performed through the fifth intercostal space. A 4 cm x 4 cm part of the native pericardium covering the left and right ventricles and the pulmonary artery were excised 1 cm anterior to the phrenic nerve. The animals were then divided randomly into the following four groups. In groups 1 (n = 5) and 2 (n = 7), the pericardial defect was closed with a patch of collagen by continuous sutures with 4-0 Prolene (Ethicon, Somerville, New Jersey), after first allowing hydration of the collagen sheet by placing it in sterile saline at room temperature for 10 minutes. In group 3 (n = 6), the pericardial defect was closed with a sterilized ePTFE (Gore-Tex pericardial membrane; W. L. Gore & Associates, Flagstaff, Arizona) patch by continuous sutures with 4-0 Prolene. In group 4 (n = 4), the pericardial defect was left as such, and no patch was used to close the defect. A thoracic drain was placed to evacuate the thoracic cavity, and the chest was closed in three layers with absorbable sutures (Vicryl; Ethicon). Subsequently, after complete suction, the thoracic drain was removed.
All the animals received ampicillin (100 mg/kg) by intramuscular injection on the day of the operation and once daily for 4 days thereafter. A median sternotomy was conducted under general anesthesia at 12 weeks after the initial operation, and the pericardial cavity was examined for adhesions, signs of infection, and the status preservation of the coronary anatomy.
For experiment 2, a total of 14 beagles with a median weight of 11.0 kg (range, 9.0 kg to 13.5 kg) were operated on, and the pericardial defect was closed with the patch used for group 1 above. All the surgical procedures were identical to those followed in experiment 1. The implanted pericardial patch was then examined at 4 weeks (n = 5), 12 weeks (n = 5), and 24 weeks (n = 4) after the first operation.
Macroscopic Evaluation
The adhesion area was determined as the area of adhesion relative to the area of the implanted sheet. The severity of the epicardial adhesions was graded on a scale of 0 to 4, and the percentage area of the pericardium abraded with adhesions was calculated, as described before [9]. The adhesions were graded for severity as follows: 0, no adhesions; 1, adhesions that could be separated easily with the fingers alone; 2, adhesions that required a combination of sharp and blunt dissection for division; 3, adhesions that required mainly sharp dissection for division. The visibility of the coronary arteries was also graded as follows: 0, clearly visible; 1, blurred; 2, completely obscured.
Microscopic Evaluation
For the light-microscopic evaluation, the specimens were fixed in 10% phosphate-buffered formalin and cut into segments containing both the central parts containing regenerated tissue and the suture line representing the zone of contact with native pericardial tissue, and a corresponding piece of native pericardial tissue as control. All the specimens were embedded in paraffin. Sections containing regenerated tissue, partially absorbed patch material, and native pericardium were stained with hematoxylin and eosin (H&E), and elastica von Gieson stain (EVG).
The light-microscopic evaluation was conducted as follows. (1) The surface lining was examined to assess the completeness of the mesotheliumlike cell layer. (2) The thickness of the regenerative tissue was evaluated in experiment 2. The thickness of the membrane was scored on a scale of 1 to 5: 1, very thin; 5, thickness of the membrane same as that of the native pericardium. If the regenerative tissue was obscure, then the thickness was estimated as the distance from the epicardium to the pleural surface of the regenerative tissue or native pericardium. (3) The amount of nonresorbed material, macrophages and lymphocytes, and the amount of collagen tissue were also estimated within the regenerated tissue. (4) Any evidence of metaplastic change was examined for, defined as an alternative differentiation of mesenchymal tissue into fat, cartilage, or bone.
Scanning Electron Microscopy
In experiment 2, the tissue specimens were fixed by immersion in 2.5% glutaraldehyde in phosphate buffer, 0.1 mol/L (pH 7.35 to 7.45). They were rinsed three times in phosphate buffer, 0.08 mol/L, before dehydration in increasing concentrations of acetone, followed by critical-point drying with carbon dioxide. Gold was sputtered on to the sections in an argon atmosphere to a thickness of 20 to 40 mm. The specimens were then examined under a scanning electron microscope.
The surface smoothness, that is, the evenness or uniformity of the surface, was evaluated. The presence of fibrin, red cells, platelets, holes within cells, and holes between cells was also assessed.
Statistics
The JMP analysis program, version 5.1 (SAS Institute, Cary, North Carolina) was used as the statistical software. All the results were expressed as mean ± SD. Intragroup comparisons of the size of the pericardial defects and the adhesion area were carried out using analysis of variance, and Dunnett's test was conducted as the post-hoc test. Analysis of the scores was performed using the corresponding nonparametric test (Kruskal-Wallis test). A p value of less than 0.05 was considered to denote significance.
|
Results |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentThe Society of Thoracic...AcknowledgmentsReferences |
|---|
Experiment 1
No significant differences in the size of the pericardial defect were found among the groups (p = 0.324). The mean pericardial adhesion area, scores, lung adhesion scores, and visibility of coronary arteries in each of the four groups are shown in Table 1.
|
|
|
In group 3, the epicardial adhesions occupied an average of 54% of the original operative area (range, 10% to 100%). Severe adhesions that were difficult to dissect were observed in 5 cases (Fig 1D). The average adhesion scores to the epicardium and lung in group 3 were 2.3 (range, 1 to 3) and 1.3 (range, 1 to 2; p = 0.567 and p = 0.013), respectively. The coronary vessels were obscured in 5 of the 6 cases, the average visibility score of the coronary vessels being 1.8 (range, 1 to 2). There were no significant differences between group 3 and group 4 in terms of the area occupied by the adhesions (p = 0.137), adhesion scores to the epicardium (p = 0.567), or the visibility score of the coronary vessels (p = 0.941); however, the two groups showed a significant difference in the adhesion score to the lungs (p = 0.013). In regard to the microscopic findings, the adhesions were composed of a thick layer of dense fibrous tissue with a population of plump fibroblastic cells surrounded by inflammatory cells and connective tissue (Fig 2C). These findings were in direct agreement with those observed at the time of macroscopic grading and were interpreted as significant host reactions to the ePTFE membrane.
In group 4, marked adhesions were observed (Fig 1E), and the retrosternal space was completely obliterated by the adhesions in all the animals. The cut pericardial edges were intimately adherent to the pleura and lung and extremely difficult to dissect out. The area of the heart exposed to the pleura and lung between the cut pericardial edges was adherent to the pleura, and the lung was covered by thick tenacious adhesions that obliterated the space, with a mean severity score of 3.0. The adhesions occupied an average of 90% of the operative area (Table 1). The adhesions in this region were so tenacious that the tissue could not be separated from the pleura or lung without damaging the myocardial tissue. The coronary vessels were severely obscured, with the mean visibility score being 2.0. Microscopically, dense adhesions were observed in the pericardial space, and these findings were in direct agreement with those observed at the time of macroscopic grading. The adhesions were made up of connective tissue with fibroblastic cells and inflammatory cells. Inflammatory reactions with occasional inflammatory cells were observed among the dense adhesions (Fig 2D).
Experiment 2
No significant differences in the size of the pericardial defect were found among the groups (p = 0.250). Macroscopically, formation of membranous tissue continuous with the pericaridium was observed in all the animals. At 4 weeks, the membranes were thin, and in 1 case the implanted collagen sheet could still be visualized at the center of the membrane (Fig 3A). At 24 weeks, the thickness score of the membranes was significantly higher than that at 4 and 12 weeks (p = 0.003), and the thickness score in each group is shown in Table 2. At 24 weeks, many new capillary vessels were observed (Fig 3B).
|
|
|
|
|
Comment |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentThe Society of Thoracic...AcknowledgmentsReferences |
|---|
Tissue injury inevitably occurs in all surgical procedures because of direct injury related to cutting, suturing and handling, or indirect injury at adjacent sites in the surgical field, related to desiccation, thermal injury, or abrasive manipulations. Histologic examination of pericardial and peritoneal tissue obtained from animals undergoing cardiac or abdominal surgical procedures indicates that damage to the mesothelium is adhesiogenic [11, 12]. Fibrin is formed in areas of pericardial mesothelial damage, and epicardial/pericardial fibrinous adhesions are initially formed. Subsequently, if these initial fibrinous adhesions are not lysed, they become organized into fibrous adhesions because of synthesis of pathologic connective tissue by activated fibroblasts. Histologic examination of peritoneal specimens from patients undergoing cardiac surgical procedures demonstrates similar initial events occurring in humans, with fibrin accumulation and inflammatory changes induced by mesothelial damage. In addition, a significant decrease of mesothelial fibrinolytic activity at the sites of injury has also been reported [13].
Many approaches have been used to reduce the risk of formation of pericardial adhesions. Pericardial substitutes have been used, and these materials could be classified into two groups: nonresorbable membranes and bioresorbable membranes.
Nonresorbable membranes include prosthetic and xenograft membranes. Prosthetic membranes include silicone rubber, ePTFE membrane, polyethylene film, and Dacron mesh [5, 14–16]. Although the ePTFE membrane has been reported to be effective, its clinical use has been limited by concerns over the permanence of this sheet and the resultant formation of a fibrous capsule. That poses a potential problem in pediatric patients as the heart grows with a foreign material in place, which could predispose to infections over time. Furthermore, ePTFE is not transparent and may interfere with visualization of the cardiac architecture during operations. Many investigators have reported on the efficacy of xenograft membranes, including equine pericardium and bovine pericardium [6, 7, 17, 18]. However, these membranes are also permanent and opaque. The current study, in which a three-layered membrane was used, demonstrates the benefits of film transparency, which allows clear visualization of the field before the sternum is closed.
Bioresorbable membranes include solutions containing pharmacologic agents and resorbable membranes. The effectiveness of several solutions, including hydrophilic polymer solution [5], hyaluronic acid coating solution [8], carboxymethyl cellulose solutions [19], and solutions containing fibrinolytic drugs, has been reported. Tissue plasminogen activator and its analogs and streptokinase have also been shown to be effective for reducing adhesion formation [20]. However, postoperative bruising, bleeding, and swelling have been reported with the use of streptokinase and tissue plasminogen activator and its analogs.
The efficacy of bioabsorbable films of polylactide [21] and polyethylene glycol and polylactic acid [22] in the prevention of adhesion formation has been reported, and polylactide has also been shown to be useful as a supporting scaffold for regeneration of the pericardium [21]. Malm and coworkers [9] studied the course of tissue regeneration after implantation of a patch made from polyhydroxybutyrate in a sheep model, and found that these patches were effective at reducing the formation of adhesions and preservation of the coronary anatomy. In the test group, light microscopy revealed the formation of a layer of mesotheliumlike cells facing the epicardial side. Recognition of such effects makes the use of bioresorbable films, which are not associated with capsule formation and do not appear to increase the long-term infection risk, more appealing.
In the present study, we have demonstrated the possibility of decreasing postoperative adhesions by closure of the pericardial sac with a collagen sheet that allows pericardial regeneration. The collagen sheet acts as a scaffold for tissue regeneration. Regeneration of the pericardium was noted in all the animals with in which the collagen sheet was implanted Formation of a complete and intact mesotheliumlike layer facing the pericardial cavity was observed in a majority of the patients. Scanning electron microscopy showed a mesotheliumlike lining completely covering the underlying collagen layer. The findings in the study confirmed that adhesions commonly do develop in dogs after cardiac procedures, as severe adhesions were observed in all the control animals.
Studies conducted using a canine model described the formation of adhesions between the epicardium and the pericardium. Unlike in the human chest, where the heart is fixed in its place by the surrounding pleura, creation of retrosternal adhesions in a canine model is difficult because of the large space between the sternum and the pericardium [22]. In this study, therefore, formation of adhesions between the posterior aspect of the sternum and the pericardium, seen in humans, was not observed. No strong adhesion formation was seen between the pericardium and the adjacent portions of the lungs, either. However, as the presence of adhesions between the posterior aspect of the sternum and the pericardium is an important problem during reoperation, it would be necessary to repeat the evaluation after establishing a more appropriate animal model.
The approaches that have been used to prevent such adhesions consist of using barrier fluids or films interposed between the injured epicardium and the sternum to physically block the formation of adhesions between the two structures. In the present study, the hyaluronic acid layer served as the barrier preventing adhesion formation and allowed regeneration of the pericardium, reducing the risks associated with resternotomy, including long-term infection, demonstrating its usefulness as a potential pericardial substitute.
In conclusion, reconstruction of the pericardium with the absorbable collagen sheet in this experiment was successful for minimizing the formation of adhesions and preserving the coronary anatomy.
|
The Society of Thoracic Surgeons Policy Action Center |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentThe Society of Thoracic...AcknowledgmentsReferences |
|---|
This website is now available at www.sts.org/takeaction.
|
Acknowledgments |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentThe Society of Thoracic...AcknowledgmentsReferences |
|---|
|
References |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentThe Society of Thoracic...AcknowledgmentsReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  J. Zhou, R. S. Liwski, C. Elson, and T. D.G. Lee Reduction in postsurgical adhesion formation after cardiac surgery in a rabbit model using N,O-carboxymethyl chitosan to block cell adherence. J. Thorac. Cardiovasc. Surg., April 1, 2008; 135(4): 777 - 783. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 I. Yoshioka, Y. Saiki, K. Sakuma, A. Iguchi, T. Moriya, Y. Ikada, and K. Tabayashi Bioabsorbable Gelatin Sheets Latticed With Polyglycolic Acid Can Eliminate Pericardial Adhesion Ann. Thorac. Surg., September 1, 2007; 84(3): 864 - 870. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ANN THORAC SURG | ASIAN CARDIOVASC THORAC ANN | EUR J CARDIOTHORAC SURG |
| J THORAC CARDIOVASC SURG | ICVTS | ALL CTSNet JOURNALS |
