Ann Thorac Surg 1998;66:1264-1268
© 1998 The Society of Thoracic Surgeons
Original articles: cardiovascular
Computed tomographic evaluation of retrosternal adhesions after pericardial substitution
Jarmo Lahtinen, MDa,
Jari Satta, MD, PhDa,
Seppo Lähde, MD, PhDb,
Ilkka Suramo, MD, PhDb,
Juha Nissinen, MDa,
Risto Pokela, MD, PhDa,
Tatu Juvonen, MD, PhDa
a Department of Surgery, Oulu University Hospital, University of Oulu, Oulu, Finland
b Department of Diagnostic Radiology, Oulu University Hospital, University of Oulu, Oulu, Finland
Accepted for publication April 27, 1998.
Address reprint requests to Dr Satta, Department of Surgery, University of Oulu, Kajaanintie 52 A, FIN-90220 Oulu, Finland
 |
Abstract
|
|---|
Background. Reoperative median sternotomy can result in cardiac injury and serious bleeding, with the rate ranging from 2% to 6%. Closure of the native pericardium can maintain a preventing plane of cleavage. In patients in whom primary pericardial closure is not possible, several substitutes have been tried with variable results. We conducted a prospective study to evaluate the clinical feasibility of polytetrafluoroethylene and polyglycolic acid patches as pericardial substitutes, using computed tomography for imaging the postoperative state of the retrosternal space.
Methods. The basic population comprised 540 patients who were scheduled for coronary artery bypass grafting, and 52 of them who met the research criteria were chosen for computed tomographic evaluation after 5 years after the primary operation.
Results. As a substitute, polytetrafluoroethylene seemed to be less adhesive to the posterior surface of the sternum. Total adhesion scores were also statistically significant (p < 0.001) to the advantage of polytetrafluoroethylene over polyglycolic acid as a pericardial substitute.
Conclusions. Polytetrafluoroethylene membrane seems to be capable of minimizing retrosternal adhesion formation and thus it may protect the heart during subsequent reoperative sternotomy.
|
Introduction
|
|---|
As the number of patients undergoing primary myocardial revascularization continues to increase, the number of potential candidates for redo operation has increased exponentially. The number of redo coronary artery bypass graftings within 10 years of the primary operation will reach 5% to 10% despite the increased use of arterial grafts [1].
Reoperations entail higher rates of morbidity and mortality, partly as a result of retrosternal adhesions, which can cause severe, even catastrophic bleeding at reentry. Therefore, certain measures should be taken at the time of the original operation to make the subsequent reopening of the sternum easier. Closure of the native pericardium can maintain a plane of cleavage between the heart and the sternum, thus helping to avoid accidental laceration of the right atrium, right ventricle, aorta, or coronary grafts at reoperation. The pericardium, however, cannot always be closed.
As an alternative to closure of the native pericardium, the pericardium can be closed with various substitutes. Several materials have been suggested and used: bovine, canine, or equine preserved pericardium, dura mater, autologous fascia lata, silicone rubber, siliconized Dacron, polypropylene and polyethylene films, or patches made of polyhydroxybutyrate [2–5]. Although several experimental studies have proved encouraging, the clinical attempts have been less satisfactory. At present, polytetrafluoroethylene (PTFE) surgical membrane and biodegradable polyglycolic acid (PGA) mesh seem to be the most suitable pericardial substitutes for closure purposes [6].
The possibilities for assessing pericardial and retrosternal adhesions before repeat operation are limited. Imaging techniques such as computed tomography (CT) can provide a detailed display of the thoracic anatomy and hence can be useful for investigating retrosternal adhesions [7–9].
The present survey was designed to evaluate the clinical feasibility of PTFE and PGA patches as pericardial substitutes, using CT for imaging the postoperative state of the retrosternal space.
|
Material and methods
|
|---|
A prospective study was conducted at the Oulu University Hospital during the period 1989 to 1994 concerning pericardial substitutes in primary isolated coronary artery bypass grafting procedures. The basic population comprised 540 patients who were scheduled for coronary artery bypass grafting, of whom 52 meeting the research criteria were chosen for CT evaluation. The inclusion criteria were as follows: age 60 years old or less; no preoperative or postoperative cortisone medication; no reoperation during primary hospital stay; no postpericardiotomy syndrome; at least one mammary artery conduit; no deep sternal infection; and mild hypothermia. The patients who entered the trial were randomly allocated into three groups by the sealed envelope method: a control group in which the pericardium was left open and allowed to retract to the sides and adhere to the mediastinal structures, a PTFE (Gore-Tex; W.L. Gore & Associates, Flagstaff, AZ) group in which the heart was covered with a thin 0.1-mm PTFE membrane anchored to the edges of the pericardium with interruptured 4-0 Prolene (Ethicon, Somerville, NJ) sutures, and a PGA group in which the heart was covered with a biodegradable PGA mesh fixed to the edges of the pericardium with interruptured 3-0 Dexon sutures. Demographic data on these groups are presented in Table 1.
A standard isolated coronary artery bypass grafting procedure with aortic cross-clamping, blood cardioplegia, and mild hypothermia (30° to 33°C) for myocardial protection was performed. Complete myocardial revascularization was aimed at using multiple venous bypass grafts and at least one internal mammary artery graft in each operation. The standardized method involved the use of two mediastinal drains, one placed intrapericardially and the other on the heart or over the surgical membrane when used.
The CTs were performed using contiguous end-inspiratory helical scanning with 5 mm collimation complemented with 1 mm high resolution scans at 20-mm intervals from the level of the ascending aorta down to the subcardiac diaphragm (GE HiSpeed Advantage; General Electric Medical Systems, Milwaukee, WI). The axial scans were filled out with midline sagittal reformats reconstructed from the overlapping helical scans (2-mm spacings). For interpretation, the images were printed on films with the mediastinal window. The same protocol was applied to 3 patients referred for chest CT for other reasons without pericardial abnormality, and these scans were used for reference purposes. The scans were evaluated by two radiologists, who first interpreted them separately and then gave a consensus interpretation. Retrosternal adhesions were considered to be present when the low-attenuating fat between the sternum and the anterior cardiac surface was absent or replaced by scar tissue. The degree or width of the adherence of scar tissue to the anterior chest wall was classified as follows: 0 = no adhesions, 1 = minimal adhesions ("peak sign") (Fig 1 ), 2 = moderate adhesions (Fig 2 ), and 3 = severe adhesions ("flat sign") (Fig 3 ). The free space between the epicardium and the substitutes was coded as follows on the grounds of the problems expected at reopening: 1 = no adherence, 2 = severe adherence. In each case the evaluation was made at three levels: the extreme cranial section was chosen at the level of the ascending aorta and main pulmonary artery, the middle level accounted for the middle portion of the right ventricle, and the extreme caudal section for evaluation was selected at the level of the acute margin of the right ventricle.
The data were stored and analyzed statistically with the SPSS for Windows statistical package, version 6.0. Means/medians and ranges are indicated. A retrosternal adhesion score, which reflects the individual degree of difficulty of adhesion formation, was calculated by summarizing the three numerical values corresponding to each level. Normality was tested first, and if it failed, the analysis was performed using the corresponding nonparametric test (Kruskall-Wallis test). Differences were considered significant at p less than 0.05.
|
Results
|
|---|
The groups were similar with regard to age, sex, preoperative New York Heart Association functional class, ejection fraction, and perioperative characteristics. There were no differences in clinical outcome during the primary hospital stay. C-reactive protein value curves were quite similar between the groups indicating that the PTFE and PGA substitutes do not cause any notable inflammatory response.
The CT evaluation showed that the middle level within each group was the most prone to adhesion formation. As a substitute, PTFE seemed to be less adhesive to the posterior surface of the sternum. In 11 of 18 patients there was only a slight scar tissue plane between the sternum and the patch (peak sign), and only 3 of 18 patients showed significant adhesions (flat sign) compared with the values in the PGA group: 3 of 17 and 9 of 17, which were quite similar to those in the control group. The key ratio, which reflects the individual degree of adhesion formation also pointed statistically to the advantage of PTFE over PGA as a pericardial substitute (p < 0.001) (Fig 4 ). Thirteen of the 18 patients (72%) in the PTFE group also had a clear free space between the epicardium and the patch compared with 6 of 17 (35%) in the PGA group. The detailed results regarding the substitutes are presented in Table 2.
View larger version (13K):
[in this window]
[in a new window]
|
Fig 4. Total adhesion scores followed by different pericardial closure methods. Higher score indicates a higher degree of adhesions found in computed tomographic follow-up (Kruskall-Wallis). (PGA = polyglycolic acid; PTFE = polytetrafluoroethylene.)
|
|
|
Comment
|
|---|
Reoperations after open heart operations are becoming more common. Resternotomies are known to result in cardiac injury in 2% to 6% of patients, and mortality rates among patients with massive bleeding are reported to range from 37% to 50% [10–12]. Complete closure of the native pericardium after cardiac operation would have the advantage of avoiding injury to the heart and the great vessels, but primary closure is often impossible, and therefore various materials have been suggested for use as a pericardial substitute to reduce the formation of adhesions. The results of using synthetic materials or glutaraldehyde-fixed pericardial xenografts have mostly been less than satisfactory, especially in the long term [13]. Good experimental animal results are generally distorted, because they did not consider the influence of the cardiopulmonary bypass itself on adhesion formation. In clinical use the impression of the suitability of the substitute is mostly suggestive being based on ease of implantation or short-term results obtained in connection with congenital heart procedures. Adequate, well-analyzed long-term data are still lacking [14]. Experimental and clinical data have nevertheless pointed to the possibility that absorbable PGA mesh and low porosity PTFE membrane may serve as acceptable methods for closing the pericardium [13, 15].
This survey supports the notion that CT, a technique relying on new technology, is appropriate and seems to serve as a good and reliable modality for evaluating the grade of retrosternal adhesion formation subsequent to open heart operation. Comparison of two well-defined pericardium substitutes revealed the superiority of expanded PTFE membrane for use in patches after cardiac operations. The convenience of the PTFE membrane relative to PGA mesh was established in the light of minor adhesion tendencies both retrosternally and between the patch and the epicardium.
The possibilities of assessing adhesions after cardiac operation are limited. A lateral chest roentgenogram to assess the presence of a retrosternal space is suggested as being somewhat indicative [6]. The parietal pericardium consists of mesothelial cells as well as a fibrous part enveloping the heart and great vessels. The pericardial sac is mostly outlined by epicardial and pericardial fat, and this permits visualization of the pericardium with CT and magnetic resonance imaging. The pericardium is visualized in 95% to 98% by CT and is best seen along the anterior aspect of the ventricular surface of the heart [16]. Duvernoy and colleagues [9], evaluating the accuracy of CT and magnetic resonance for imaging postoperative adhesions subsequent to cardiac operation and before reoperation, found that CT was superior to magnetic resonance imaging for detecting retrosternal adhesions. The findings at CT were in accordance with those found at operation regarding postoperative retrosternal adhesions in the cranial retrosternal space in 14 of 15 patients, and in the caudal retrosternal space in 12 of 15 patients. This possibility of visualizing the retrosternal fat replaced by scar tissue with CT is in accordance with the findings of Silverman and Harell [16], who considered that the normal pericardium is best visualized on CT in sections through the ventricles, where epicardial and pericardial fat is most abundant. Intrapericardial adhesions could not be detected responsively by CT, however. These previous results, indicating that CT evaluation and operative findings go hand in hand regarding retrosternal adhesions, allow us to assume that our results will also be of clinical value when planning cardiac reoperations.
The number of actual reoperations involving patients with previously implanted PTFE pericardial substitutes is gradually increasing, and experiences are mostly encouraging relative to those acquired with other previously used substitutes. Harada and colleagues [2] used PTFE as a pericardial substitute in 61 children with congenital heart disease, 23 of whom underwent reoperation. The membrane could be removed easily, and the anterior surface of the heart beneath it was covered with a thin layer of fibrous tissue that obscured the fine surface features but allowed visualization of the principal cardiac structures. Jacobs and colleagues [15] have recently conducted a multicenter observational program on the use of PTFE, inserting this membrane in 1,085 patients. Reoperations were performed safely and effectively on 105 patients. No problematic adhesions were observed between the membrane and the heart or between the membrane and the chest wall and the rate of injury at resternotomy (1%) was low compared with those previously reported in the literature.
The low adhesive response of the PTFE material may be attributable to its small pore size, which prevents cellular penetration and therefore, provides a safe plane for dissection while inducing a minimum amount of reaction and adhesions. Conversely, absorbable PGA mesh has large pores, which may permit free migration of cells and nutrients through it. This helps us understand the contribution of exchange to adhesion formation. Furthermore the absorbative property of PGA induces a pronounced fibroproliferative reaction, and hence also a dense adhesion zone around it.
In conclusion, by CT evaluation the PTFE membrane seemed to be less adhesive to the posterior surface of the sternum compared to PGA, thus offering the possibility of avoiding cardiac injury during subsequent reoperative sternotomy. Insertion of the PTFE membrane is not a substitute for primary closure of the pericardium, however, or of meticulous surgical techniques. The variable findings suggestive of adhesions obtained by CT indicate a need for further studies to assess the role of the latter in the preoperative assessment of retrosternal adhesions and to predict possible difficulties that are postponed when planning redo cardiac operation.
|
References
|
|---|
- Sergeant P., Lesaffre E., Flameng E., Suy R. Internal mammary artery: methods of use and their effect on survival after coronary bypass surgery. Eur J Cardiothorac Surg 1990;4:72-78.[Abstract]
- Harada Y., Imai Y., Kurosawa H., Hoshino S., Nakano K. Long-term results of the clinical use of an expanded polytetrafluoroethylene surgical membrane as a pericardial substitute. J Thorac Cardiovasc Surg 1988;96:811-815.[Abstract]
- Minale C., Nikol S., Hollweg G., Mittermayer C., Messmer B.J. Clinical experience with expanded polytetrafluoroethylene Gore-Tex surgical membrane for pericardial closures: a study of 110 cases. J Card Surg 1988;3:193-201.[Medline]
- Eng J., Ravichandran P.S., Abott C.R., Kay P.H., Murday J.A., Shreiti I. Reoperation after pericardial closure with bovine pericardium. Ann Thorac Surg 1989;48:813-815.[Abstract]
- Malm T., Bowald S., Bylock A., Busch C. Prevention of postoperative pericardial adhesiones by closure of the pericardium with absorbable polymer patches. J Thorac Cardiovasc Surg 1992;104:600-607.[Abstract]
- Loop F.D. Catastrophic hemorrhage during sternal reentry. Ann Thorac Surg 1984;37:271-272.[Medline]
- Silverman P.M., Harell G.S. Computed tomography of the normal pericardium. Invest Radiol 1983;18:141-144.[Medline]
- Doppman J.L., Reinmuller R., Lissner J., et al. Computed tomography in constrictive pericardial disease. J Comput Assist Tomogr 1981;5:1-11.[Medline]
- Duvernoy O., Malm T., Thuomas K.
., Larsson S.G., Hansson H.E. CT and MR evaluation of pericardial and retrosternal adhesions after cardiac surgery. J Comput Assist Tomogr 1991;15:555-560.[Medline]
- Macmanus Q., Okies J.E., Phillips S.J., Starr A. Surgical considerations in patients undergoing repeat median sternotomy. J Thorac Cardiovasc Surg 1975;69:138-143.[Abstract]
- Dobell A.R.C., Jain A.K. Catastrophic hemorrhage during redo sternotomy. Ann Thorac Surg 1984;37:273-278.[Abstract]
- Wideman F.E., Blackstone E.H., Kirklin J.W., Karp R.B., Kouchoukos N.T., Pacifico A.D. Hospital mortality of replacement of the aortic valve: incremental risk factors. J Thorac Cardiovasc Surg 1981;82:692-698.[Abstract]
- Gabbay S., Guindy A.M., Andrews J.F., Amato J.J., Seaver P., Khan M.Y. New outlook on pericardial substitution after open heart operations. Ann Thorac Surg 1989;48:803-812.[Abstract]
- Bogers J.C. Massive bleeding at resternotomy despite a polytetrafluoroethylene surgical membrane. J Thorac Cardiovasc Surg 1995;110:274-275.[Free Full Text]
- Jacobs J.P., Iyer R.S., Weston J.S., et al. Expanded PTFE membrane to prevent cardiac injury during resternotomy for congenital heart disease. Ann Thorac Surg 1996;62:1778-1782.[Abstract/Free Full Text]
- Silverman P.M., Harell G.E. Computed tomography of the normal pericardium. Invest Radiol 1983;18:141-144.
This article has been cited by other articles:
|
|
|
|
 
K. Saha, W. McEwan, F. A Frizelle, and H. Singh
Limiting Pericardial Adhesions after Coronary Artery Bypass: Experimental Study
Asian Cardiovasc Thorac Ann,
March 1, 2004;
12(1):
61 - 64.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
M. Katayama, T. Masui, S. Kobayashi, T. Ito, H. Sakahara, A. Nozaki, and H. Kabasawa
Evaluation of Pelvic Adhesions Using Multiphase and Multislice MR Imaging with Kinematic Display
Am. J. Roentgenol.,
July 1, 2001;
177(1):
107 - 110.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
P. Leprince, M. Rahmati, N. Bonnet, V. Bors, A. Rama, Ph. Leger, I. Gandjbakhch, and A. Pavie
Expanded polytetrafluoroethylene membranes to wrap surfaces of circulatory support devices in patients undergoing bridge to heart transplantation
Eur. J. Cardiothorac. Surg.,
March 1, 2001;
19(3):
302 - 306.
[Abstract]
[Full Text]
[PDF]
|
|
|
Top
Abstract
Introduction
