Pericardial substitutes: a survey

Outcomes of expanded polytetrafluoroethylene pericardial membrane implantation in left ventricular assist device explantation and heart transplantation

OBJECTIVES

Redo sternotomy and explantation of left ventricular assist devices (LVAD) for heart transplantation (HT) involve prolonged dissection, potential injury to mediastinal structures and/or bleeding. Our study compared a complete expanded polytetrafluoroethylene (ePTFE) wrap versus minimal or no ePTFE during LVAD implantation, on outcomes of subsequent HT.

METHODS

Between July 2005 and July 2018, 84 patients underwent a LVAD implant and later underwent HT. Thirty patients received a complete ePTFE wrap during LVAD implantation (Group 1), and 54 patients received either a sheet of ePTFE placed in the anterior mediastinum or no ePTFE (Group 2).

RESULTS

Baseline characteristics were similar between Groups 1 and 2. Surgeons reported subjective improvements in speed, predictability, and safety of dissection with complete ePTFE compared with minimal or no ePTFE. Time from incision to initiation of cardiopulmonary bypass (CPB) were similar between groups (97 ± 38 vs. 89 ± 29 min, p = .3). Injury to mediastinal structures during the dissection was similar between groups (10% vs. 11%, p > .9). While surgeons reported less intraoperative bleeding in Group 1 (43% vs. 61%), this trend did not reach significance (p = .1). In-hospital mortality, intensive care unit length of stay and hospital length of stay were similar between both groups.

CONCLUSIONS

In patients undergoing LVAD explant-HT, there was a trend toward reduced surgeon reported intraoperative bleeding with ePTFE placement. Despite qualitatively reported greater ease and speed of mediastinal dissection with ePTFE membrane placement, time to initiation of CPB did not differ, likely because surgeons remained cautious, allowing extra time for unanticipated difficulties.

Management of retrosternal adhesion after median sternotomy by controlling degradation speed of a dextran and ε-poly (L-lysine)-based biocompatible glue

OBJECTIVE

Retrosternal adhesion after median sternotomy possibly raises the risk of cardiac injury at resternotomy. A biodegradable glue "Lydex" is composed of food additives, dextran and ε-poly (L-lysine), and the degradation speed can be controlled by the composition. In the present study, we evaluated the preventative effect of Lydex on retrosternal adhesion and the relationship between degradation speed and the progression of retrosternal fibrosis.

METHODS

Japanese white rabbits are subjected to median sternotomy. Lydex 1, 2 and 3 were loaded at the retrosternal space of rabbits in allocated groups before sternal closure, respectively (n = 11 for each group). Retrosternal adhesion was macroscopically evaluated after surgery. Retainment of Lydex, retrosternal fibrosis and the infiltration of macrophages are histologically evaluated, respectively.

RESULTS

All Lydex groups exhibited less retrosternal adhesion at 4 weeks after loading compared to unloaded control. The degradation speed of Lydex varied according to the compositions. Lydex with faster degradation (Lydex 2 or Lydex 3) showed lower progression of retrosternal fibrosis compared to that with slower degradation (Lydex 1) [fibrosis ratio: control vs Lydex 1 vs Lydex 2 vs Lydex 3: 0.60 ± 0.15 vs 0.18 ± 0.17 vs 0.00 ± 0.00 vs 0.00 ± 0.00, P = 0.0005 (Lydex 1 vs Lydex 2), P = 0.0005 (Lydex 1 vs Lydex 3)]. Retrosternal infiltrations of macrophages in Lydex 1 and Lydex 3 groups are not higher compared to that in unloaded control.

CONCLUSIONS

The degradation speed of Lydex could be controlled according to the compositions. The degradation speed affected the progression of retrosternal fibrosis.

Suspected involvement of EPTFE membrane in sterile intrathoracic abscess and pericardial empyema in a multi-allergic LVAD recipient: a case report

Device-related infections in recipients of left ventricular assist devices (LVAD) have been recognized as a major source of morbidity and mortality. They require a high level of diagnostic effort as part of the overall burden resulting from infectious complications in LVAD recipients. We present a multi-allergic patient who was treated for persistent sterile intrathoracic abscess formation and pericardial empyema following minimally invasive LVAD implantation including use of a sheet of e-polytetrafluoroethylene (ePTFE) membrane to restore pericardial integrity. Sterile abscess formation and pericardial empyema recurred after surgical removal until the ePTFE membrane was removed, suggesting that in disposed patients, ePTFE may be related to sterile abscess formation or sterile empyema.

Alternative for Primary Pericardial Closure: Sentry for Re-entry

BACKGROUND

Redo cardiac surgery is considered high-risk surgery as accidental injury to the aorta, the innominate vein, the ventricles and the atria is a possibility. Such accidental injury occurs when the cardiac chamber is adherent to the undersurface of the sternum. Closure of pericardium at the time of primary surgery can prevent adherence of cardiac chambers to the sternum, but may increase the risk of tamponade. This study aimed to show that covering heart with a pedicled pericardial fat pad not only serves the purpose of cover but also avoids the adverse haemodynamic effects of primary pericardial closure.

METHODS

Forty patients undergoing elective cardiac surgery were randomised into two groups depending on the way pericardium was managed. Both techniques were already in routine use in our unit and in other units around the country. One method is to leave the pericardium widely open, the other is to loosely oppose the pericardial fat pad over the surface of the aorta and right ventricle. Twenty-three patients had a pedicled pericardial fat pad covering the heart: Closure Group. Seventeen patients had no pericardial fat pad cover over the heart: Open Group. A haemostasis clip was used as a radio-opaque marker over the epicardium in both groups. Post-operation heart rate, central venous pressure, pulmonary artery diastolic pressure, mean arterial pressure and cardiac index were measured and recorded 1, 3 and 8h after surgery. The distance between the haemoclip and the posterior table of the sternum was measured at 6 days and 6 months post-operation. Haemodynamic parameters and the retrosternal space depth were compared between the two groups.

RESULTS

There were no important differences in haemodynamic parameters between the two groups. Post-operative lateral chest Roentgenograms showed that the distance between epicardial surface and the posterior table of sternum was larger in the Closure Group compared to Open Group on post-operative day 6, 17.5+/-1.0mm versus 13.4+/-1.3mm (P=0.0013) and 6 months later, 12.3+/-0.8mm versus 6.0+/-1.2mm (P<0.001). There was no mortality in either group.

CONCLUSION

Pedicled pericardial fat pad cover is a good alternative to primary pericardial closure as there are no adverse haemodynamic effects in early post-operative period and the long-term benefit of protection of heart at the time of re-sternotomy can be expected.

Expanded polytetrafluoroethylene membrane for prevention of adhesions in patients undergoing external decompression and subsequent cranioplasty

Cranioplasty performed after external decompression for brain swelling may be difficult because of the development of adhesions between the temporal muscle and the dura. Membranes composed of expanded polytetrafluoroethylene (ePTFE) were inserted at external decompression to prevent such adhesions. The ePTFE membranes were placed suturelessly between the temporal muscle and the dura, covering the proximal portion of the temporal muscle at the sphenoidal ridge and the dural suture line. In addition, part of the membrane was placed directly below the skin incision to allow easy visualization at subsequent cranioplasty in 10 patients. Cranioplasty was performed 38 to 126 days (mean 63.9 days) after external decompression. No adhesions were observed, either between the ePTFE membranes and the dura, or between the membranes and the temporal muscle. No temporal muscle and dural injuries occurred during cranioplasty, and no patient showed signs of infection. Mean operating time was 71.0 minutes and mean blood loss was 75.2 ml. Sutureless insertion of ePTFE membranes at external decompression increases the ease and safety of subsequent cranioplasty, decreases operating time and blood loss, and avoids injury to the temporal muscle and dura.

Expanded polytetrafluoroethylene membranes to wrap surfaces of circulatory support devices in patients undergoing bridge to heart transplantation

OBJECTIVE

Because of a lack of donor hearts, an increasing number of patients with heart failure must now undergo bridge to cardiac transplantation with a mechanical circulatory support device. Moreover, support periods have become longer. As a result, pericardial adhesions may develop while the support device is implanted, increasing the risk of injury at resternotomy and bleeding after transplantation. Use of expanded polytetrafluoroethylene (ePTFE) pericardial substitutes (membranes) may prevent such adhesions.

PATIENTS AND METHODS

From January 1997 to December 1999, ePTFE membranes were used in 23 patients to wrap portions of an implanted left ventricular assist device (LVAD) or total artificial heart (TAH). Any complications during mechanical support or at cardiac transplantation were recorded. Six ePTFE membranes removed at transplantation were studied histologically.

RESULTS AND CONCLUSIONS

At resternotomy for transplantation, the plane of dissection between tissues, ePTFE membranes, and surfaces of the mechanical support device were easily discerned. No adhesions were observed between tissues and membranes. There were no injuries during resternotomy and no patient had to undergo reoperation because of bleeding. One patient given a TAH had an infection during circulatory support that was controlled by antibiotic therapy. In another patient, clots developed between the device and an ePTFE membrane; these were removed successfully. Histologic studies of removed ePTFE membranes showed no infiltration of the membranes interstices by collagen or cellular components. Use of ePTFE membranes in patients undergoing bridge to transplantation with either an LVAD or a TAH limited adhesions between tissues and device surfaces without increasing the risk of infection.

Should the pericardium be closed routinely after heart operations?

BACKGROUND

Repeat coronary artery bypass grafting is more difficult if the right ventricle is firmly attached to the inner table of the sternum. Closure of the pericardium at the time of the initial procedure may prevent attachment of the right ventricle to the sternum. This study attempts to identify the geometric effects of pericardial closure early after isolated coronary artery bypass grafting.

METHODS

Forty-two patients undergoing elective, isolated coronary artery bypass grafting were randomized into two groups: 20 patients underwent closure of the pericardium (Closure group) and the pericardium was left open in 22 patients (Open group). Radiopaque markers were attached to the anterior aspect of the right ventricular epicardium in both groups.

RESULTS

Postoperative chest roentgenograms revealed that the distance between the epicardial surface and the posterior table of the sternum was larger in the Closure group compared to the Open group at 1 week and 3 months postoperatively (p < 0.001). Cardiac index and stroke work index in the early postoperative period was lower in the Closure group compared to the Open group (p < 0.001) despite similar filling pressures.

CONCLUSIONS

Pericardial closure may reduce the risk of myocardial injury during sternotomy for repeat coronary artery bypass grafting by preventing right ventricular adhesions. However, adverse hemodynamic effects in the early postoperative period may preclude pericardial closure in patients with impaired ventricular function.

Prevention of retrosternal adhesion formation in a rabbit model using bioresorbable films of polyethylene glycol and polylactic acid

The purpose of this study was to test the efficacy of three bioresorbable films of polyethylene glycol (EO) and polylactic acid (LA) (EO/LA = 1.5, 2.5, and 3.0) in the prevention of adhesion formation between the epicardium and the sternum (retrosternal adhesions) in a rabbit model. Retrosternal adhesions were generated by sternotomy, pericardiotomy, and abrasion of the anterior epicardium. The adhesion barrier was placed between the epicardium and the sternum and sutured to the edge of the pericardium. Epicardial adhesions were evaluated 14-20 days later by assessing the area of the epicardium covered by adhesions. In the control rabbits, tenacious adhesions were observed between sternum and the central portion of epicardium (portion exposed through the pericardiotomy) which were difficult to dissect. When a bioresorbable film was placed over the pericardium, adhesion formation at the central strip of the epicardium (area between the sternum and the epicardium exposed through the pericardium) could be reduced or prevented. At this site, the areas of adhesion formation were 0% (EO/LA = 1.5), 8.4 +/- 2.8% (EO/LA = 2.5), and 5.6 +/- 4.7% (EO/LA = 3.0) of the central strip, significantly less than that observed in the control group, 78.0 +/- 5.8% (P < 0.01). At the anterior left and right and posterior apex of the heart (sites where the film was not placed), there were no differences between control and treatment groups. The films were completely resorbed at the time of necropsy in group EO/LA = 2.5 and 3.0. Small pieces of film were observed in group EO/LA = 1.5. In conclusion, the bioresorbable films [EO/LA = 1.5 (REPEL-CV), 2.5, or 3.0] were efficacious in the reduction of retrosternal adhesions to the epicardium.

Expanded PTFE membrane to prevent cardiac injury during resternotomy for congenital heart disease

BACKGROUND

Resternotomy for repair of congenital cardiac defects can result in cardiac injury. Closure of the pericardium during the initial operation may prevent this, and several pericardial substitutes have been tried, with variable results, in patients in whom primary pericardial closure is not possible. We conducted a multicenter observational study of the use of the expanded polytetrafluoroethylene membrane (Preclude Pericardial Membrane, formerly called the Gore-Tex Surgical Membrane; W. L. Gore & Associates, Flagstaff, AZ) in patients likely to undergo reoperation for treatment of congenital heart disease.

METHODS

Data were collected retrospectively on all patients in whom the expanded polytetrafluoroethylene membrane was inserted at the initial operation for congenital heart disease at 12 centers in 1984 to 1993.

RESULTS

A total of 1,085 patients (mean age, 55 +/- 2.5 months) received the membrane. During follow-up ranging from 1.3 to 10.5 years, 105 reoperations were performed. Injury during resternotomy occurred in only 1 patient (1% of reoperations). There were no membrane-related deaths or complications in the entire series of 1,085 patients.

CONCLUSIONS

The expanded polytetrafluoroethylene membrane was safe and effective in helping to prevent cardiac injury during resternotomy for treatment of congenital heart disease.

Technique to protect the internal mammary artery pedicle

A simple technique to protect the internal mammary artery pedicle from injury in reoperative sternotomies is described. It consists of a pericardial window created by two pericardial flaps with displacement of the mammary pedicle away from the midline.

Protection of the internal mammary artery pedicle with polytetrafluoroethylene membrane

With increasing frequency, reoperative coronary artery bypass surgery is being performed in the setting of a patent internal mammary artery (IMA) graft. Injury to the IMA graft at reoperation can result in ischemic myocardial injury and cardiac arrest. This descriptive laboratory study examined use of a polytetrafluoroethylene (PTFE) membrane wrap to protect the IMA pedicle during sternal reentry. Six pigs (25-30 kg) underwent median sternotomy and takedown of left and right IMAs. Grafts were implanted in the epicardium by a modified Vineberg procedure. In each animal, one IMA was wrapped circumferentially with PTFE, while the contralateral bare IMA served as the control. Redo sternotomy was performed 3 months later. Severe adhesions between chest wall, myocardium, and bare IMA grafts were encountered in six of six animals. PTFE-wrapped IMAs were easily identified by appearance and by palpation. There was no adherence between PTFE membranes and surrounding tissue, nor was there evidence of cellular infiltration, disruption, or contraction of the membrane. Most notably, PTFE membranes could not be cut or injured with the electrocautery, since they are electrical nonconductors. This study demonstrates that PTFE membrane: (1) is relatively inert and incites minimal surrounding tissue reaction; and (2) is impenetrable to electrocautery injury. Further studies are necessary to determine the impact of PTFE membrane on IMA graft patency and wound infection. PTFE membrane may prove clinically useful in protecting patent IMA pedicles during reoperative cardiac surgery.

Pericardial closure without pericardial substitute

Primary closure of the native pericardium is recommended whenever possible, and pericardial substitutes are advocated when primary closure is not feasible. To avoid foreign material, we have employed a unique method of dissection to allow mediastinal coverage without tension.

Autologous pericardial flap for prevention of reentry injury in cardiac reoperations

Cardiac injury during sternal reentry to the heart is a rare but dangerous complication of cardiac reoperations. Positioning a pericardial flap between the heart and sternum at the time of the initial operation may consistently facilitate cardiac reoperation by providing a reliable plane of dissection and by reducing adhesion formation.

Use of expanded polytetrafluoroethylene pericardial substitute with ventricular assist devices

Expanded polytetrafluoroethylene membrane was used to reconstruct the pericardium after ventricular assist device insertion in 7 consecutive patients. One patient remains on support with the membrane in place and 6 have had the membrane removed. The expanded polytetrafluoroethylene membrane protected the device cannulas during repeat sternotomy and expedited the dissection of the heart and great vessels at the time of device removal. Reconstruction of the pericardium with expanded polytetrafluoroethylene membrane is recommended at the time of ventricular assist device insertion.

The Gore-Tex Surgical Membrane: effectiveness as a barrier to inhibit postradical pelvic surgery adhesions in a porcine model

Radical pelvic surgery has well-defined side effects, many of which may be related to the formation of pelvic adhesions. The effectiveness of different agents at limiting the formation of postradical pelvic surgery adhesions (PRPSA) has been variable to date. Use of a barrier could be beneficial not only by limiting adhesion formation but by acting as a "pelvic lid" to elevate the bowel out of a radiation treatment field and therefore limit subsequent radiation injury and associated fistula formation. We investigated the ability of the Gore-Tex Surgical Membrane (Gore-SM) to inhibit PRPSA in 20 adult female hogs undergoing radical hysterectomy, bilateral salpingo-oophorectomy, and resection of pelvic peritoneum. After completion of the radical resection, animals were randomized to either no attempt at covering the peritoneal defect or covering it and a 1-cm margin of intact peritoneum with a tailored sheet of Gore-SM. The membrane was secured in place using a continuous running 4-O Prolene suture around the perimeter. No intraoperative deaths occurred. No animal evidenced signs of bleeding, infection, bowel obstruction, or abscess formation. Four weeks after the initial surgery, the animals were again anesthetized, exploratory celiotomy was performed, and adhesions were quantified with specific note being made of any segments of small bowel that were adherent into the pelvis. Animals were then sacrificed. Adhesion scores for the Gore-SM-treated animals (n = 10; mean, 0.14 +/- 0.12; median, 0.21) were significantly less than those of animals with the noncovered pelvis (n = 10; mean, 1.33 +/- 0.41; median, 1.42; P less than 0.001). Similarly, significantly fewer animals treated with Gore-SM had small bowel loops adherent in the pelvis when compared to control animals (10% vs 70%; P less than 0.01). In this model, meticulously suturing the membrane in place with continuous permanent suture, the Gore-SM was an effective barrier for postoperative adhesion prophylaxis and successfully limited small bowel adherence into the pelvis.

The inability of Gore-Tex Surgical Membrane to inhibit post-radical pelvic surgery adhesions in the dog model

We investigated the ability of Gore-Tex Surgical Membrane (Gore-SM) to inhibit PRPSA formation in 10 adult female canines undergoing radical hysterectomy, bilateral salpingo-oophorectomy, omentectomy, and resection of the pelvic/abdominal peritoneum. At the time of surgery, one-half of the pelvic/abdominal peritoneal defect was covered with a tailored single layer of the Gore-SM. The membrane was sutured into place with 3-O Dexon in an interrupted manner. Each animal served as its own internal control. No operative/postoperative deaths or postoperative complications occurred. Four weeks postsurgery, euthanasia was effected, necropsy performed, and adhesions quantified. Adhesion scores for the Gore-SM-covered areas (n = 10; mean score, 2.76 +/- 1.47; median, 2.46) were significantly higher than those for control areas (n = 10; mean score, 1.46 +/- 2.13; median, 1.81; P = 0.01). Adhesions to the Gore-SM occurred at wrinkles in or at the edges of the membrane. In contradistinction to the findings of other investigators using different animal models, the Gore-SM appears to increase PRPSA in this unique model.

A new biological membrane for pericardial closure

In an attempt to develop a new and better biological membrane for the pericardium, we evaluated the use of glutaraldehyde treated human amniotic membrane (AM) and compared it to polytetraflouroethylene (PTFE) membrane as a pericardial substitute. Four dogs underwent a right thoracotomy. Two 4 x 3-cm portions of pericardium, 3-4 cm apart, were excised in each animal and both sites replaced with a similar sized patch of 0.8% glutaraldehyde-treated AM and 0.2-mm PTFE membrane respectively. The AM was sutured to the pericardial defect with its smooth surface facing the epicardium. After 18 weeks, the materials were evaluated. The AM patches revealed minimal extrapericardial and no epicardial adhesion. The PTFE patches had moderate to severe epicardial adhesions and severe extrapericardial adhesions. Histopathological examination of AM patches revealed that the integrity of the AM was preserved, however, neovascularization and slight lymphocytic infiltration were observed. In the PTFE patches, there were severe inflammatory infiltration and fibroblast proliferation into both surfaces. AM patches by virtue of their low immunogenicity evoke very minimal host to graft reaction. These AM grafts, however, tear easily unless carefully sutured. Improved methods of treatment may improve suturing strength. We conclude that AM performs satisfactorily and is superior to PTFE as a pericardial substitute.

Scanning electron microscopic observations of cell populations and bacterial contamination of membranes used for guided periodontal tissue regeneration in humans

Fourteen intrabony periodontal defects and six furcation defects (Class II) were treated by a flap procedure which included the use of a polytetrafluoroethylene membrane to allow guided tissue regeneration. After 4 to 6 weeks of healing, the membranes were retrieved and examined by scanning electron microscopy for the presence of adherent cells and other tissue elements. The cervical open pore-structured collar of the membrane, which in most cases had become partially exposed to the oral cavity, had a deposit of bacterial plaque. Bacterial Bacterial colonies and a scatter of single cells in some instances extended into the mid-third of the membrane. Fibroblast-like cells and, in some specimens, blood vessels and fibrous structures were seen in the mid-third and deep parts of the membrane. Generally, however, the occlusive portion of the membrane was characterized by a sparseness of adherent tissue elements. There did not seem to be a systematic difference in the nature and distribution of the adherent structures on the inner and outer surfaces of the membrane. The findings suggest that, in addition to preventing flap tissues from contacting the root surface, an important function of the membrane is to protect the integrity of the underlying blood clot by diverting mechanical stress acting on the flap during early stages of healing.

Reoperation after pericardial closure with bovine pericardium

To reduce the risk of damage to the anterior surface of the right ventricle after resternotomy, it has been recommended that the pericardium be closed with a patch after open heart operations. We have examined 4 patients undergoing resternotomy for the third time 3 to 8 years after bovine pericardium valve replacement. On each occasion the pericardium was closed with a patch of bovine pericardium. In all cases, the patch was frozen to the inner aspect of the sternum, increasing the difficulty of resternotomy. Histological examinations of the patches confirmed dense fibrous connective tissue, patchy calcification, and foreign body giant cell reaction. Bovine pericardium appears to increase the difficulty of repeat cardiac operations. We recommend its use be discontinued.