HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Add to Personal Folders Download to citation manager Author home page(s): Paul R. Vogt Marko I. Turina Permission Requests Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by von Segesser, L. K. Articles by Turina, M. I. Search for Related Content PubMed PubMed Citation Articles by von Segesser, L. K. Articles by Turina, M. I. Related Collections Related Article

Ann Thorac Surg 1995;60:511-515
© 1995 The Society of Thoracic Surgeons

---

Original Articles: Cardiovascular

Prevention of Residual Ventricular Septal Defects With Fibrin Sealant

Clinics for Cardiovascular Surgery and Pediatric Cardiology, University Hospital, Zürich, Switzerland

	Abstract

Top Footnotes Abstract Introduction Material and Methods Results Comment References

 
Background. Modern echocardiography now allows for the detection of a substantial number of residual ventricular septal defects (VSDs) after surgical patch repair that remained hidden in the past. Mostly without hemodynamic significance, residual VSDs may have clinical consequences (progressive dehiscence, hemolysis, prophylactic antibiotic treatment, endocarditis). To reduce the number and size of residual VSDs we performed an experimental and a clinical study.

Methods. (1) In an experimental setup, burst pressure of 60 fibrin glue-sealed defects (calibrated between 1.0 and 5.0 mm in diameter) was determined using a computerized recording system and pressure loads up to 500 mm Hg. (2) In a prospective clinical trial with blinded postoperative echocardiographic controls VSD closure was performed in 36 consecutive patients (age, 37 ± 40 months; range, 4 to 134 months) using a polytetrafluoroethylene patch and running sutures reinforced with pledgets (22 of 36 patients) or sealed with fibrin glue (14 of 36 patients) in accordance to the surgeon's preference.

Results. (1) Experimentally, mean pressure load achieved was more than 500 ± 0 mm Hg for 1.0-mm defects, 413 ± 52 mm Hg for 2.5-mm defects, 363 ± 58 mm Hg for 4.0-mm defects, and 313 ± 48 mm Hg for 5.0-mm defects (r 0.873, p < 0.001). (2) Clinically, all patients survived. Residual VSDs at echocardiography were observed in 16 of 22 patients (72%) for reinforced versus 5 of 14 patients (36%) for sealed with fibrin glue (p < 0.05). Diameter of residual VSDs accounted for 1.3 ± 1.2 mm for reinforced versus 0.3 ± 0.4 mm for sealed with fibrin glue (p < 0.01). Hemodynamically significant residual VSDs were found in 2 of 22 patients (9%) for reinforced versus 0 of 14 patients (0%) for sealed with fibrin glue (p = not significant).

Conclusions. Small defects sealed with fibrin glue resist physiologic pressure load. Fibrin glue sealing of prosthetic patches during intracardiac VSD repair allows for significant reduction of number and size of residual VSDs. Improved long-term outcome can be expected.

	Introduction

Top Footnotes Abstract Introduction Material and Methods Results Comment References

 
See also page 515.

Surgical patch closure of ventricular septal defects (VSDs) is now well established. However, almost any larger series reports on some hemodynamically significant residual VSDs after surgical repair [1–3]. Furthermore modern echocardiography [4, 5] allows for detection of a substantial number of residual VSDs that remained hidden in the past. Although most of these residual VSDs are without hemodynamic significance, this diagnosis may have clinical consequences, as in some series of infective endocarditis, a VSD was the main underlying heart disease [6]. On the other hand, it is well accepted that surgical closure of VSDs lowers the risk of bacterial endocarditis [7]. Interestingly, the size of the VSD is not associated with the risk of bacterial endocarditis [7]. Hence, lifelong antibiotic prophylaxis of infective endocarditis remains indicated even in very small residual VSDs.

Two types of problems can be considered responsible for residual VSDs. On one hand, there is the well-known friability of the myocardial tissue that does not always allow for secure anchoring of the sutures. This is particularly evident for zones where superficial suturing is recommended to avoid disturbance of the atrioventricular conduction system. Of course, pull-through of sutures is particularly problematic with running sutures as it can result in a loose suture line. On the other hand, the diagnosis of a loose patch can also be attributable to ``unsutured'' segments of the patch circumference, which in turn may be due to incongruences between the prosthetic patch and the underlying myocardial structures, inadequate exposure of the circumference of the VSD, or combinations.

A number of techniques to improve the strength of the sutures has been suggested in the past including Teflon pledgets, strips of Teflon felt, or strips of pericardium. Although these techniques help to reduce the number of sutures pulled through the muscular septum they are not always suitable to occlude geometric incongruences between septum and patch. In an effort to reduce size and number of residual VSD after surgical patch closure we evaluated the use of fibrin glue, which is well known for its occlusive properties from extracardiac [8, 9] as well as intracardiac applications [10, 11]. Before clinical application we evaluated the potential strength of fibrin glue in vitro.

	Material and Methods

Top Footnotes Abstract Introduction Material and Methods Results Comment References

 
Experimental Evaluation
To determine the sealing potential of fibrin glue, we studied in vitro various defects occluded with fibrin glue. Sixty silicone rubber tubing segments (outer diameter, 9.2 mm; inner diameter, 6.4 mm; wall thickness, 1.4 mm) were perforated laterally with an aortic punch. Four basic punch sizes (diameters: 1.0, 2.5, 4.0, and 5.0 mm) were used to produce circular defects. Fibrin glue (Tissucol, Immuno AG, Zürich, Switzerland) was used as thermoinactivated preparation of the two main components, human fibrinogen and human thrombin (the latter was previously provided from a bovine source), which are now available fully prepared in frozen form. After thawing, the premounted and filled double syringes are ready for use. Only the mixing connector and a blunt needle have to be added.

Burst pressure of defects sealed with fibrin glue was determined in sequential fashion: (1) Calibration of the punched defect with a probe. Only circular defects with known diameter were included. (2) Sealing of the lateral defect with fibrin glue at room temperature. (3) Connection of the tubing segment (with the sealed lateral defect) to a computerized recording system on one side and a line providing compressed air to the other. (4) Three minutes after sealing, progressive pressure load from 0 mm Hg up to 500 mm Hg was applied. (5) Analysis of the pressure loading curve.

Loading pressures were read directly from a calibrated pressure transducer with digital read out (Veri-Cal, Utah Medical Products, Midvale, UT), whereas burst pressures were determined by a microtip pressure transducer (Millar, Houston, TX) and transferred to a computerized recording system (Dasa, Gould Inc, Cleveland, OH) allowing for graphic and mathematic analyses.

Clinical Study
Clinical application was studied in a prospective trial using for VSD closure a polytetrafluoroethylene patch and running sutures that were either reinforced with pledgets or sealed with fibrin glue and blinded postoperative echocardiographic controls.

Thirty-six consecutive patients (mean age, 37 ± 40 months; range, 4 to 134 months) undergoing VSD patch closure were studied. In accordance to the surgeon's preference, 22 patients had VSD patch repair with reinforced sutures and 14 patients with sealed patches. For the group operated on with reinforced sutures, the diagnoses included an isolated VSD in 8 of 22 patients (36%) as compared to a VSD in Fallot, double-outlet right ventricle, and more complex situations in 14 patients (64%). In contrast, for the group with sealed patches, there was only one isolated VSD (1 of 14 patients, 7%; p = 0.05) as compared to VSDs in tetralogy of Fallot, double-outlet right ventricle, and more complex situations in 13 patients (93%; p = 0.05). Preoperative shunt fraction was 52 ± 17% for reinforced sutures as compared to 72 ± 16% for sealed patches (p < 0.01).

With exception of the type of VSD closure, all procedures were performed in standardized fashion using microporous hollow fiber oxygenators (Lilliput, Masterflow or Midiflow, Dideco Spa, Mirandola, Italy), clear priming (whenever possible), and moderate hypothermia (24° to 28°C). Repair of the VSD and left-sided lesions of the heart were performed with cold blood cardioplegia-induced cardiac arrest, whereas right-sided lesions were repaired during rewarming with fibrillating heart and open aorta. Transatrial, transpulmonary, or transaortic approach for VSD closure was selected whenever possible. Transventricular approach was only used in cases with double-outlet right ventricle or pulmonary atresia type right outflow tract. Transatrial VSD closure was performed in 16 of 22 patients (73%) for reinforced sutures as compared to 8 of 14 patients (57%) for sealed with glue (p = not significant [NS]). Transpulmonary approach was selected in 3 of 22 patients (14%) for reinforced sutures as compared to 2 of 14 patients (14%) for sealed with glue (p = NS). Transaortic approach was used in 1 of 22 patients (5%) for reinforced sutures as compared to 1 of 14 patients (7%) for sealed with glue (p = NS). Transventricular route was selected in 2 of 22 patients (9%) for reinforced sutures as compared to 3 of 14 patients (21%) for sealed with glue (p = NS).

Application of Fibrin Sealant
Fibrin glue is now available as deep frozen preparation (see Experimental Evaluation), which is thawed during cannulation, cardioplegia application, and inspection of the VSD. As outlined above, transatrial approach through the tricuspid valve is preferred. Figure 1 shows the operative exposure and repair of a perimembranous VSD using an expanded polytetrafluoroethylene patch (0.4- or 0.6-mm thickness; W. L. Gore & Associates Inc, Flagstaff, AZ). The patch is attached with a loose running 6/0 expanded polytetrafluoroethylene suture (W. L. Gore). The inset of Figure 1 shows that some fibrin glue is applied behind the patch before the suture is tightened (eg, with a hook) and the patch is seated on a bed of liquid glue. A second layer of glue is applied on the border of the patch, the sutures, and the ventricular septum nearby. A speedy technique is necessary as the fibrin glue coagulates up to some degree within seconds. After 60 s the glue becomes reasonably solid. To achieve adequate adherence, the surfaces to be sealed have to be dried before application of the glue. Hence, we use a vent on the left side of the heart (through the interatrial septum), a pump sucker on the right side, and swabs in the peripatch area to keep the blood out of the zone to be glued. Rubbing of the endothelial contact area with a sponge provides superior adherence. Care must be taken not to soil the tricuspid valve and its subvalvar apparatus as well as the cavities of the left heart where embolization could occur.

View larger version (61K): [in this window] [in a new window]   Fig 1. . Transatrial approach to the repair of a perimembranous ventricular septal defect demonstrating the surgical technique. The borders of the ventricular septal defect patch are sealed with fibrin glue in two steps. The patch is seated on a layer of fibrin glue before the suture is tightened. An additional layer of glue is added after (see inset). Caveat: region to be sealed must be dry to allow for acceptable adhesion and soiling of left cavities and tricuspid valve with fibrin glue have to be avoided.

Statistical Analyses
In the experimental study, mean ± standard deviation was derived for all pressure loads as well as for the respective groups corresponding to the different defects. Systat statistical software package (Systat Inc, Evanston, IL) was used for calculation of linear regression.

For the clinical study, mean ± standard deviation was calculated for parametric data and Student's t test was used where applicable for comparison between groups. Fisher`s exact test was used for comparison of nonparametric data (Solo, distributed by BMDP Statistical Software, Los Angeles, CA).

Statistical significance was confirmed by a p value of less than 0.05.

	Results

Top Footnotes Abstract Introduction Material and Methods Results Comment References

 
Experimental Evaluation
Compilation of all data points (n = 60) resulted in a mean pressure load of 438 ± 78 mm Hg. The number of test runs per size of defect was 7 of 60 for 5.0-mm diameter, 8 of 60 for 4.0-mm diameter, 15 of 60 for 2.5-mm diameter, and 30 of 60 for 1.0-mm diameter. Pressure loads achieved per group were 313 ± 48 mm Hg for 5.0-mm defects, 363 ± 58 mm Hg for 4.0-mm defects, 413 ± 52 mm Hg for 2.5-mm defects, and 500 ± 0 mm Hg for 1.0-mm defects. Linear regression analysis of pressure loads achieved estimates the intercept (constant) with 543 and the slope with (-47) and the correlation is (multiple r) 0.873 (p < 0.01).

All 1.0-mm defects sealed with fibrin glue resisted the selected maximal pressure load of 500 mm Hg. For bursted seals, the recorded pressures were 196 ± 129 mm Hg for 2.5-mm defects, 113 ± 61 mm Hg for 4.0-mm defects, and 63 ± 52 mm Hg for 5.0-mm defects (Fig 2). Linear regression analysis of the burst pressures estimates the intercept (constant) with 591 and the slope with (- 120). The correlation between diameter and burst pressure (multiple R) is 0.890 (p < 0.001).

View larger version (24K): [in this window] [in a new window]   Fig 2. . Experimental evaluation. Mean ± standard deviation of pressures achieved in bursted seals. Linear regression analysis provided a multiple r value of 0.890 (p < 0.001).

The postoperative results are summarized in Figure 3. Residual VSDs at echocardiography were observed in 16 of 22 patients (72%) for reinforced sutures as compared to 5 of 14 (36%) for sealed with fibrin glue (p < 0.05). Diameter of residual VSDs accounted for 1.3 ± 1.2 mm (range, 0 to 4 mm) for reinforced sutures as compared to 0.3 ± 0.4 mm (range, 0 to 1 mm) for sealed with fibrin glue (p < 0.01). Hemodynamically significant residual VSDs were found in 2 of 22 patients (9%) for reinforced sutures as compared to 0 (0%) for sealed with fibrin glue (p = NS).

View larger version (25K): [in this window] [in a new window]   Fig 3. . Clinical study. Postoperative results are given as percent per group analyzed. There are significantly fewer residual ventricular septal defects (VSDs) in the group with sealed patches (p < 0.05).

	Comment

Top Footnotes Abstract Introduction Material and Methods Results Comment References

In our hands, application of these techniques to the clinical setting (ie, fibrin glue sealing of prosthetic patches during intracardiac VSD repair) allowed for significant reduction of number and size of residual VSDs despite more complex procedures performed in the study group. As outlined before, the fibrin sealant can compensate for small incongruences between cardiac structures and patch material. This function is similar to the naturally occurring clots sealing vascular anastomoses, which in this case are provoked deliberately before the exposure to the bloodstream. Furthermore, if well placed, the fibrin glue can, due to its adherence, also contribute to better patch fixation (improved distribution of forces) and therefore, reduce the risk of suture pull- through. This function is similar to the pledgets used in the control group. In contrast to additional stitches, however, application of fibrin glue will not interfere with the conduction system.

The adequate resistance of glue deposited in the interventricular septum has also be demonstrated by Leca and colleagues [11] who reported on surgical treatment of multiple VSDs using biological glue. Her group studied the use of fibrin glue for repair of artificially created VSDs in sheep with sizes between 5 and 10 mm and achieved satisfactory results. Interestingly, necropsy studies showed that the glue was progressively replaced with fibrous tissue. In the clinical study reported by the same group, fibrin glue was used mainly for closure of apical VSDs and proved to be very reliable. Three-year follow-up further confirmed that repair of VSDs with fibrin glue remained intact over time as suggested by the previous experimental work. The recommendation to use fibrin glue for closure of muscular VSDs with less than 8 mm of diameter is in agreement with our finding that sealed defects of 5 mm or less in diameter can hold considerable pressure.

Potential drawbacks of glue application within the heart include soiling of structures that should not be included in the area to be sealed (eg, subvalvar apparatus of the tricuspid valve) as well as embolization of glue particles deposited on areas with a blood film where proper adhesion is not possible. Both problems can be avoided if adequate exposure and proper techniques are applied as outlined above. Venting of the left-sided cavities should be nonocclusive during the sealing procedure to avoid aspiration of glue.

Despite the application of fibrin glue on a large scale, there are only few reports on side effects. Even immunization by bovine thrombin [12] used in the mixture should not occur anymore as thrombin of human origin is now used in the frozen preparations. At this time, the only bovine component still necessary is aprotinin (3,000 KIU/mL).

Routine application of fibrin glue and its potential side effects have to be balanced against the risk related to residual VSDs of minor size on one side and the risk of reoperation in hemodynamically significant VSDs on the other. Traditionally, a VSD was mainly considered for surgical closure if pulmonary-to-systemic flow ratio was more than 2.0 and in general the same criteria were applied for closure of residual VSDs after surgical patch closure. Recently, Backer and colleagues [13] reported on closure of restrictive ventricular septal defects with pulmonary-to-systemic flow ratio of less than 2.0. Despite excellent results in their series this issue remains controversial as no real long-term results are available yet [14, 15]. However, it is a fact that a significant number of reoperations have to be performed after surgical patch closure of VSD even if only the so-called hemodynamically significant shunts are considered. Castaneda and colleagues [16] found two significant shunts after VSD closure in a series of 146 cases early after operation. Another three significant residual left-to-right shunts were found during late follow-up bringing the rate of significant residual VSDs up to 5 of 148 or 3.4%. Admittedly there was also an unknown proportion of so-called trivial residual VSDs at cardiac catheterization. Unmentioned is the proportion of residual VSDs that are only detectable with modern echocardiographic techniques. The former and the latter may be without importance with regard to the development of pulmonary artery hypertension. However, lifelong antibiotic prophylaxis remains indicated if the VSD was not completely closed during operation [7]. Even higher proportions of residual VSDs were reported for patients with multiple VSDs. Kirklin and co-workers [17] observed a reoperation rate of 28%. Again only hemodynamically significant VSDs were reoperated and there is an unknown proportion of residual VSDs with a shunt fraction of less than 2.0. Furthermore the proportion of residual VSDs detectable only at echocardiography is not available. In contrast to these findings, there was no reoperation in the series of Leca and colleagues [11] who used fibrin glue to close small VSDs in patients with multiple defects.

Considering the importance of reliable complete closure of VSDs, we believe that sealing of the prosthetic patch with fibrin glue during surgical repair is a practical way to reduce the number and size of residual VSDs. With proper technique the potential risks of this approach are minimal and improved long-term outcome can be expected.

	Footnotes

Top Footnotes Abstract Introduction Material and Methods Results Comment References

 
Presented at the Thirty-first Annual Meeting of The Society of Thoracic Surgeons, Palm Springs, CA, Jan 30–Feb 1, 1995.

Address reprint requests to Dr von Segesser, Clinic for Cardiovascular Surgery, University Hospital, CH-8091 Zürich, Switzerland.

	References

Top Footnotes Abstract Introduction Material and Methods Results Comment References

 

1. Sairanen H, Leijala M, Louhimo I. Surgery for ventricular septal defect. Scand J Thorac Cardiovasc Surg 1991;25:1–5.[Medline]
2. Vogt J, Wesselhoeft H, Luig H, et al. The pre- and postoperative findings in 627 patients with tetralogy of Fallot. Thorac Cardiovasc Surg 1984;32:234–43.[Medline]
3. Bical O, Perrier P, Fermont L, et al. Réopération après correction chirurgicale de tetralogie de Fallot. Arch Mal Coeur-Vaiss 1984;77:595–9.
4. Andrade JL, Serino W, de Leval M, Somerville J. Two dimensional echocardiographic assessment of surgically closed ventricular septal defects. Am J Cardiol 1983;52:325–9.[Medline]
5. Roberson DA, Muhiudeen IA, Cahalan MK, et al. Intraoperative transesophageal echocardiography of ventricular septal defect. Echocardiography 1991;8:687–97.[Medline]
6. Fukushige J, Igarashi H, Ueda K. Spectrum of infective endocarditis during infancy and childhood: 20 year review. Pediatr Cardiol 1994;15:127–31.[Medline]
7. Gersony WM, Hayes CJ, Driscoll DJ, et al. Bacterial endocarditis in patients with aortic stenosis, pulmonary stenosis, or ventricular septal defect. Circulation 1993;87(Suppl 1):121–6.[Medline]
8. Seelich T. Tissucol (Immuno Vienna): biochemistry and methods of application. J Head Neck Pathol 1982;3:65–9.
9. Rousou J, Levitsky S, Gonzales-Lavin L, et al. Randomized clinical trial of fibrin sealant in patients undergoing resternotomy or reoperation after cardiac operations. J Thorac Cardiovasc Surg 1989;97:194–203.[Abstract]
10. Von Segesser LK, Oechslin E, Jenni R, Turina MI. Use of glue to avoid formation of perfused recesses in aortic homograft implantation. Ann Thorac Surg 1994;57:494–5.[Abstract]
11. Leca F, Karam J, Vouhé P, et al. Surgical treatment of multiple ventricular septal defects using biologic glue. J Thorac Cardiovasc Surg 1994;107:96–102.[Abstract/Free Full Text]
12. Berruyer M, Amiral J, French P, et al. Immunization by bovine thrombin used with fibrin glue during cardiovascular operations. J Thorac Cardiovasc Surg 1993;105:892–7.[Abstract]
13. Backer CL, Winters RC, Zales VR, et al. Restrictive ventricular septal defect: how small is too small to close. Ann Thorac Surg 1993;56:1014–9.[Abstract]
14. Malm JR. A reason to close ventricular septal defect? Ann Thorac Surg 1993;56:1013.[Medline]
15. Waldman JD. Why not to close a small ventricular septal defect? Ann Thorac Surg1993;56:1011–2.[Medline]
16. Castaneda A, Jonas RA, Mayer JE, Hanley FL. Cardiac surgery of the neonate and infant. Philadelphia: Saunders, 1994:187–201.
17. Kirklin JK, Castaneda AR, Keane JF, Fellows KE, Norwood WI. Surgical management of multiple ventricular septal defects. J Thorac Cardiovasc Surg 1980;80:485–93.[Abstract]

This article has been cited by other articles:

				A. Haussler and R. Pretre Surgical closure of a perimembranous ventricular septum defect with a running suture MMCTS, May 23, 2008; 2008(0523): 2410. [Abstract] [Full Text] [PDF]

				S. A. LeMaire, S. A. Carter, T. Won, X. Wang, L. D. Conklin, and J. S. Coselli The Threat of Adhesive Embolization: BioGlue Leaks Through Needle Holes in Aortic Tissue and Prosthetic Grafts Ann. Thorac. Surg., July 1, 2005; 80(1): 106 - 111. [Abstract] [Full Text] [PDF]

				A. F. Merrick, M. Lal, R. H. Anderson, and D. F. Shore Management of ventricular septal defect: a survey of practice in the United Kingdom Ann. Thorac. Surg., September 1, 1999; 68(3): 983 - 988. [Abstract] [Full Text] [PDF]

This Article Abstract Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Add to Personal Folders Download to citation manager Author home page(s): Paul R. Vogt Marko I. Turina Permission Requests Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by von Segesser, L. K. Articles by Turina, M. I. Search for Related Content PubMed PubMed Citation Articles by von Segesser, L. K. Articles by Turina, M. I. Related Collections Related Article