| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Ann Thorac Surg 2005;80:1835-1840
© 2005 The Society of Thoracic Surgeons
a Department of Cardiovascular Surgery, Graduate School of Medicine, Tohoku University, Sendai, Japan
b Department of Medical Electronics, Suzuka University of Medical Science, Sendai, Japan
Accepted for publication April 27, 2005.
* Address correspondence to Dr Sakuma, Department of Cardiovascular Surgery, Graduate School of Medicine, Tohoku University, 1-1 Seiryomachi Aoba-ku, Sendai 980-8574, Japan (Email: ksakuma{at}mail.tains.tohoku.ac.jp).
 |
Abstract |
|---|
 Top Abstract IntroductionMaterial and MethodsResultsCommentReferences |
|---|
-caprolactone copolymer (L-C copolymer). In addition, e-PTFE sheet and autologous pericardium were used as controls. METHODS: Sixty dogs were divided into four groups of 15. In group A, a 3 x 3 cm segment of pericardium was excised, and the autologous pericardium was resutured. In group B, the pericardial defect was replaced with gelatin sheet. In group C, the defect was replaced with L-C copolymer sheet. In group D, the defect was replaced with e-PTFE sheet. For each group, the implanted membranes were retrieved at 2 weeks (n = 1), 4 weeks (n = 3), 12 weeks (n = 5), and 24 weeks (n = 6) after implantation.
RESULTS: The e-PTFE sheet produced severe adhesions to the heart and pleura and a more prominent inflammatory reaction, as compared with the gelatin sheet. The absorbable pericardial substitutes were completely absorbed by 24 weeks after implantation, and were replaced with fibrous membrane.
CONCLUSIONS: Gelatin sheet may involve less adhesion and a reduced inflammatory reaction compared with e-PTFE.
|
Introduction |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentReferences |
|---|
Several materials can be used as pericardial substitutes, including silicone rubber [1], polyurethane [2], fascia lata [3, 4], expanded-polytetrafluoroethylene (e-PTFE) sheet [2, 5, 6], heterologous porcine, equine, or bovine pericardium [7–10], Dacron [11, 12], and dura mater [13]. However, several drawbacks have been reported with the use of pericardial substitutes—including dense adhesions and severe inflammatory reactions. At reoperation, these complications give rise to longer operative and perfusion times, bleeding, and injury to the heart or great vessels. And, occasionally, the use of foreign material for pericardial substitutes causes mediastinitis that emerges as a profound complication after open heart surgery. Recently, many authors have described materials that reduce the formation of adhesion in the retrosternal space [14, 15]. These materials, however, are not yet widely used because of possible problems associated with their clinical application. On the other hand, Yamada and associates [16] described in 1997 a composite bioabsorbable sheet, composed of two copolymer films of L-lactic acid--caprolactone (50% L-lactic acid, 50% -caprolactone) and a poly(glycolic acid) nonwoven fabric, which displayed good mechanical properties, was completely absorbed by 24 weeks after implantation in the back of rats, and was replaced with the regenerated duralike tissue in dural defects in rabbits. They concluded that this new bioabsorbable composite sheet can be successfully used as a dural substitute.
The present study was designed to investigate the absorption rate and the tissue reaction to two absorbable pericardial substitutes: gelatin sheet [24] and L-lactic acid--caprolactone copolymer (L-C copolymer). In addition, the most commonly used e-PTFE sheet and autologous pericardium were used as controls.
|
Material and Methods |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentReferences |
|---|
Animal care was in compliance with the Principles of Animal Use formulated by the Animal Use Committee at Tohoku University.
Experimental Groups and Pericardial Substitutes
The animals were divided into four groups of 15 dogs each. In group A, were the controls (a segment of pericardium was excised, and the autologous pericardium was resutured). In group B, the pericardial defect was replaced with a 0.1-mm thick gelatin sheet (Gunze, Kyoto, Japan). In group C, the pericardial defect was replaced with a 0.4-mm thick L-lactic acid--caprolactone (50:50) copolymer sheet (Gunze). In group D, the defect was replaced with 0.1-mm thick expanded-polytetrafluoroethylene (E-PTFE) sheet (W.L. Gore & Associates, Elkton, Maryland).
The fabrication processes for the gelatin sheet and L-C copolymer (50:50) sheet are described below.
Gelatin Sheet
Gelatin, extracted from bovine bone (type I collagen) using the alkaline method, was donated by Nitta Gelatin (Osaka, Japan). To prepare gelatin films, gelatin powder was dissolved in distilled water to a concentration of 10 wt%. The solution was cast on glass plates and allowed to dry in air, yielding gelatin films of 0.1 mm in thickness. Cross-linking was introduced by exposing both sides of the gelatin films to ultraviolet light for 10 hours, in air, at 60 cm from a 15-W ultraviolet lamp (Toshiba GL-15). The sheet was ready for use as a pericardial substitute after sterilization with ethylene oxide gas [17].
L-Lactic Acid--Caprolactone (50/50) Copolymer Sheet (50:50) Sponge Sheet
The L-C copolymer (50:50 mol%, Mw = 220,000) was dissolved in 1,4-dioxane to a concentration of 6 wt%. This solution was poured into a mold and was frozen at –135°C. Then, the frozen content was freeze-dried to yield a porous sponge sheet of 0.5-mm thickness. This sponge sheet was completely dried at 70°C for 12 hours in vacuo. The sheet was ready for use as a pericardial substitute after sterilization with ethylene oxide gas [17].
Surgical Procedure
An intravenous line and electrocardiographic monitoring were established. Anesthesia was induced with 2.5% sodium thiopental at a dose of 25 mg/kg, and maintained with halothane in oxygen. Mechanical ventilation was instituted with an approximate tidal volume set at 10 mL/kg body weight, with 100% oxygen, at a rate of 8 to 14 cycles per minute. Under sterile conditions, a left lateral thoracotomy incision was made through the fourth or fifth intercostal space. After pericardial fat and feeding vessels were carefully separated and the surface of the pericardium was exposed, a 3 x 3 cm segment of pericardium was excised about 1 cm anterior to the phrenic nerve. After scattering 5 mL of autologous blood into the pericardial defect (in order to promote the formation of adhesions), a sheet was anchored to the surrounding pericardium with 5-0 Prolene (Ethicon, Somerville, New Jersey) mattress sutures. Care was taken to avoid contact with the epicardial surface of the heart. Prophylactic antibiotics (cefazolin sodium 40 mg/kg) were given intravenously just after the operation.
Dogs were sacrificed at 2 to 24 weeks after operation and the en bloc heart, including the pericardium and left lung, was removed through a left-sided thoracotomy. The implant tissue sites were then examined and photographed.
Evaluations
The survival rates for the 4 groups were 100% (group A), 93% (group B), 93% (group C), and 100% (group D). The implanted membranes were retrieved (from each group) at 2 weeks (n = 1), 4 weeks (n = 3), 12 weeks (n = 5), and 24 weeks (n = 6) after implantation. Adhesion formation was evaluated by macroscopic findings—particularly in the pleura or the lung-to-pericardium (P) and pericardium-to-epicardium (E) interfaces. The degree of adhesion was quantitatively classified from 0 to 3, using the adhesion score of Heydorn and colleagues [18]: 0 = no adhesion; 1 = adhesion could be readily separated by finger dissection; 2 = adhesion is of intermediate strength; and 3 = adhesion necessitated sharp dissection.
In each animal, specimens of myocardium, epicardium, and pericardium were taken for light microscopic studies. These were fixed in 10% phosphate-buffered formalin for at least 3 days, embedded in paraffin, and sectioned at 5 µm. Then, sections were stained with hematoxylin-eosin and elastica-Masson in order to assess inflammatory reaction as well as changes in the pericardial substitutes themselves.
Escalating severity of the inflammatory reaction was histopathologically defined by increasing numbers of infiltrating inflammatory cells and inflammatory foci. The inflammatory reaction of each sheet was quantitatively classified from 0 to 3, using the inflammatory reaction score reported by Lu and colleagues [19]: 0 = no cell infiltration; 1 = sparse, focal infiltration of neutrophils, lymphocytes, and plasma cells; 2 = focal infiltration of neutrophils, plasma cells, and lymphocytes; 3 = diffuse infiltration of neutrophils, plasma cells, and lymphocytes. The fibrous reaction was quantitatively classified from 0 to 3, using the fibrous reaction score reported by Lu and coworkers [19]: 0 = no fibrous reaction; 1 = sparse, focal fibrous connective tissue, hyalinization, and fibrin deposition; 2 = a thin layer of fibrous connective tissue, hyalinization, and fibrin deposition; and 3 = a thick layer of fibrous connective tissue, hyalinization, and fibrin deposition. Adhesion, inflammatory reaction, and fibrous reaction were evaluated by the same two pathologists, who were masked to the treatment. Quoted scores were the mean of the two scores classified by the pathologists.
Finally, the fibrous membrane was observed using cytokeratin immunohistochemistry.
Statistical Methods
Results are expressed as mean ± SD. Analysis of variance (ANOVA) and the Scheffe test for post-hoc comparison were used to evaluate differences among the four groups. A p value of less than 0.05 was considered to be significant for the test.
|
Results |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentReferences |
|---|
After 2 weeks, there were no adhesions of either pleura-to-pericardium or pericardium-to-epicardium, and no differentiation between the four groups.
After 4 weeks, group A demonstrated minimal adhesion of the autologous pericardium, and the adhesions of both pleura-to-pericardium and pericardium-to-epicardium could be separated by finger in 1 dog. However, these findings were not consistent. In another 2 animals, autologous pericardium had moderately or completely adhered to the underlying epicardium. In group B, a thin membrane had replaced the gelatin sheet with no adhesions within the pericardium, in all animals, and the membrane was covered with autologous tissue and lung. In group C, the membrane had incompletely replaced the original caprolactone sheet; the degree of epicardial adhesion was none in 1 dog and minimal to moderate in the other 2 dogs. In group D, 2 dogs had no adhesions between the pleura and the e-PTFE seat or the between e-PTFE seat and the epicardium, but dense adhesion was observed in 1 dog.
After 12 weeks, group A adhesions were absent in 2 dogs and moderate in 3 dogs (between the pericardium and epicardium). Adhesions between the pericardium and pleura were minimal in 3 dogs and moderate in 2 dogs. In group B, the pericardial space was free of adhesions in 4 of 5 dogs and there were no adhesions between the pleura and pericardium in 4 of 5 dogs. One dog had minimal adhesion in both interfaces. In group C, 2 dogs had moderate adhesion between the pericardium and epicardium, and 3 had none. Adhesion between the pericardium and pleura was minimal in 3 and moderate in 2. In group D, dense adhesion was seen between the e-PTFE sheet and the epicardium in 1 of 5 dogs, while adhesion was moderate in 3 dogs and minimal in 1.
After 24 weeks, in group A, adhesion between the pericardium and epicardium was absent in 1 dog, minimal in 3 dogs, and tight in 2 dogs; adhesion between the pericardium and pleura was absent in 1 dog, minimal in 2 dogs, moderate in 2 dogs, and tight in 1 dog. In group B, the pericardial space was free of adhesions in 5 of 6 dogs, and no dog had adhesions between the pleura and pericardium. In group C, the original caprolactone sheet was completely replaced by a thin membrane in 4 of 6 dogs, but only incompletely in 2 dogs. Adhesion between the pericardium and epicardium was absent in 3 dogs, minimal in 1, and moderate in 2. Adhesion between the pericardium and pleura was absent in 2 dogs, minimal in 2, and moderate in 2. In group D, dense adhesion was seen between the e-PTFE sheet and the epicardium in 4 of 6 dogs; adhesion here was moderate in 1 dog and absent in the remaining dog.
The macroscopic results are summarized in Tables 1 and 2
(numbers represent mean ± SD of the adhesion scores). Adhesion formation on the gelatin sheet, both the pleural side and the pericardial side, was significantly less than on the e-PTFE sheet, at 12 and 24 weeks after operation (p < 0.05).
|
|
|
|
|
|
|
|
|
|
Comment |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentReferences |
|---|
Experimental results in animals have been far superior to the results of human clinical trials [5, 7, 20], which have demonstrated varying degrees of success. The reason for this difference is unclear, but Gabbay and colleagues [14] pointed out that it may be explained by blood clotting after cardiopulmonary bypass.
Of the synthetic sheets, those made from e-PTFE are superior to the others [5] and have been widely used as pericardial substitutes. The e-PTFE sheet has, however, a number of significant disadvantages; it remains in situ as a permanent foreign body and causes extensive inflammatory reaction. The use of such sheets might increase the incidence of mediastinitis [23]. Therefore, it was of interest to evaluate the host reaction to absorbable pericardial substitutes (gelatin sheet and L-C copolymer) in this study.
Moderate adhesion was observed between the pleura and the bioabsorbable materials at 4 weeks after implantation of the absorbable pericardial substitutes. However, the newly formed membranous tissue could be easily removed at 12 and 24 weeks after operation. Once bioabsorbable materials were totally absorbed, no host reaction remained [22].
Although the difference was not statistically significant for the L-C copolymer sheet, both absorbable pericardial substitutes had less adhesion than did the autologous pericardium and the e-PTFE sheet, a finding consistent with the report by Okuyama and associates [21]. When fibrous reaction causes adhesion, the first stage is the adherence of fibroblasts to the pericardial substitute. Then, fibroblasts proliferate and produce collagen fibers. There have been reports that the nature of the adhesion of cells to a material is related to the water content of the material [25]. Kulik and associates [25] reported that nonionic hydrogels exhibited minimal platelet adhesion at a water content of around 90%. The adhesion of fibroblasts may also be related to the water content of the material.
Microscopic examinations also revealed that whereas a strong foreign body reaction was present at 4 weeks after implantation, the number of infiltrating inflammatory cells was substantially reduced at 12 and 24 weeks after the operation. The gelatin sheet had less adhesion formation than did the L-C copolymer sheet. The extent of adhesion formation can be, at least in part, explained by the difference in the absorption rate. While the gelatin sheet needed 4 weeks to be totally absorbed, the L-C copolymer sheet needed 12 weeks. It is speculated that there is some relationship between the absorption rate and the degree of adhesion formation, but adhesion formation may be more closely related to the characteristics of the bioabsorbable materials.
In this study, where the section of tissue was replaced with a bioabsorbable pericardial substitute, a layer of the cells stained positively for cytokeratin and was likely mesothelium. Further investigation may be required to explain the origin of these cells.
There are several reports on the use of L-C copolymer sheet to repair defects in the dura mater, in experimental models. Among these reports, Yamada and coworkers [16] have reported that an L-C copolymer sheet is gradually absorbed and replaced by a dura-like connective tissue. In our study, bioabsorbable sheets were completely absorbed and replaced by fibrous membranous tissue. The gelatin sheet had less fibrous reaction than did the L-C copolymer sheet. The fibrous membranous tissue that replaced the L-C copolymer sheet had adequate strength, while only a very thin membrane was formed upon absorption of the gelatin sheet. Complete closure of the pericardial defect with membranous tissue formed after absorption of the pericardial substitute may adequately protect the heart from injury at the time of resternotomy, but avoidance of adhesion formation is a far more important component. The gelatin sheet is, therefore, still promising for clinical application as a pericardial substitute.
In conclusion, both the gelatin sheet and the L-C copolymer sheet were completely absorbed by 24 weeks after implantation. The gelatin sheet exhibited reduced formation of pleural and pericardial adhesions as compared with the e-PTFE sheet, at 12 and 24 weeks after implantation. Thus, gelatin sheet should be useful for reconstruction as an absorbable substitute. In the medium and late phases, gelatin sheets had permitted only slight adherence, and their complete absorption meant that no foreign body remained.
|
References |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsCommentReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  G. Matsumura, T. Shin'oka, Y. Ikada, T. Sakamoto, and H. Kurosawa Novel Anti-Adhesive Pericardial Substitute for Multistage Cardiac Surgery Asian Cardiovasc Thorac Ann, August 1, 2008; 16(4): 309 - 312. [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]
|
|
|
|
 |
|
 S. Masuda, Y. Saiki, S. Kawatsu, I. Yoshioka, H. Fujiwara, S. Kawamoto, S. Sai, A. Iguchi, N. Sakamoto, T. Ohashi, et al. Trial of new vascular clips for aortic anastomosis in a canine model J. Thorac. Cardiovasc. Surg., September 1, 2007; 134(3): 723 - 730. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Y. Chang, P.-H. Lai, C.-C. Wang, S.-C. Chen, W.-C. Chang, and H.-W. Sung Mesothelium regeneration on acellular bovine pericardia loaded with an angiogenic agent (ginsenoside Rg1) successfully reduces postsurgical pericardial adhesions J. Thorac. Cardiovasc. Surg., October 1, 2006; 132(4): 867 - 874. [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 |
