|
|
||||||||
Ann Thorac Surg 2005;80:106-111
© 2005 The Society of Thoracic Surgeons
The Texas Heart Institute at St. Lukes Episcopal Hospital and The Michael E. DeBakey Department of Surgery, Division of Cardiothoracic Surgery, Baylor College of Medicine, Houston, Texas
Accepted for publication February 1, 2005.
* Address reprint requests to Dr LeMaire, One Baylor Plaza, BCM 390, Houston, TX 77030 (Email: slemaire{at}bcm.tmc.edu).
Presented at the Fortieth Annual Meeting of The Society of Thoracic Surgeons, San Antonio, TX, Jan 2628, 2004.
| Abstract |
|---|
|
|
|---|
METHODS: Polypropylene suture was used to create end-to-end anastomoses in gelatin-sealed woven polyester grafts (n = 45), expanded polytetrafluoroethylene (ePTFE) grafts (n = 45), and fresh porcine aortas (n = 45). An additional 45 anastomoses were created in ePTFE grafts using ePTFE sutures. The outer surface of each anastomosis was covered with BioGlue. Anastomoses underwent inspection with direct magnification or histology.
RESULTS: BioGlue leaked through needle holes and into the lumen in 10% of anastomoses (18 of 180). Leaks were significantly more common in fresh aorta (10 of 45, 22%) than in prosthetic grafts (8 of 135, 6%; p = 0.003). Suture size did not significantly affect the incidence of leak. The use of ePTFE sutures did not eliminate BioGlue leakage. Prosthetic graft leaks created discreet round adhesive particles. In contrast, aortic tissue leaks resulted in thin, friable flakes of glue extending along the intimal surface. Aortic histology confirmed that BioGlue reached the vessel lumen via the suture channels.
CONCLUSIONS: BioGlue leaked through the needle holes in fresh aortic tissue and prosthetic grafts. Intraluminal adhesive particles were easily dislodged, supporting concerns regarding embolization. The potential for adhesive embolization should be a factor when considering the relative risks and benefits of using BioGlue.
| Introduction |
|---|
|
|
|---|
Several authors [35] have found BioGlue to be effective in improving hemostasis during thoracic aortic surgery, especially during cases of acute proximal aortic dissection. While the use of BioGlue to optimize hemostasis is promising, it raises important safety concerns. We have previously documented [810] that direct contact with BioGlue is harmful to exposed nerves and cardiac conduction tissue and that BioGlue interferes with aortic growth. Other reports [1115] have raised concerns that intraluminal fragments of glue may embolize. Carrel and colleagues [12] have speculated that adhesives may gain access to the vessel lumen through suture-line needle holes. The purpose of this study was to determine if BioGlue leaks through anastomotic needle holes in aortic tissue and two common prosthetic graft materials.
| Material and Methods |
|---|
|
|
|---|
Microscopic Examination
The anastomoses in the porcine aortas were divided into segments (1 cm x 0.6 cm), fixed in 4% paraformaldehyde at room temperature for 18 hours, and embedded in paraffin. Sections were cut (5 µm) along the vessels longitudinal plane, perpendicular to the anastomosis, and stained with hematoxylin and eosin.
Statistical Analysis
Statistical analysis was performed to determine the difference in leakage rates between groups. We analyzed the differences using two-tailed Fishers exact tests and p < 0.05 was regarded as statistically significant. All analyses were performed with SPSS for Windows release 11.0.1 (SPSS Inc, Chicago, IL).
| Results |
|---|
|
|
|---|
|
|
|
| Comment |
|---|
|
|
|---|
The potential threat of adhesive embolization was articulated by von Segesser and colleagues [11] in their 1995 report regarding a series of patients who underwent ventricular septal defect closure using a patch and fibrin sealant. While there were no strokes in these patients, the authors emphasized the importance of avoiding spillage of sealant into the left ventricle in order to prevent embolization. That same year, Carrel and colleagues [12] reported a case of surgical adhesive embolization. A patient with acute proximal aortic dissection underwent surgical repair of the ascending aorta and transverse aortic arch using a polyester graft, polypropylene suture, felt strips, and GRF glue. The patient died 3 weeks later due to sepsis. Although the patient did not have clinical evidence of stroke, postmortem examination revealed multiple cerebral infarctions and polymerized glue particles within cerebral arteries. The authors theorized 3 mechanisms for embolization: (1) direct spillage of glue into the true lumen (despite precautions); (2) escape of glue through distal reentry sites into the true lumen; and (3) secondary mobilization of glue particles through suture-line needle holes.
Several subsequent reports have confirmed that adhesives can inadvertently enter the systemic circulation and can embolize. Gillham and Tousignant [18] described a case of aortic mechanical valve occlusion caused by GRF glue that appeared to have entered the aortic lumen through the aortotomy suture-line. Similarly, Birmingham [19] reported finding fibrin sealant on an immobilized disk of a mechanical aortic valve prosthesis. Gallet and colleagues [20] reported systemic embolization of glue after endoscopic obturation of gastroesphogeal varicies with glue. Miller [13] described extracting large BioGlue "bullets" from the thoracoabdominal aorta in patients who had undergone previous proximal aortic repairs. Guerrero and colleagues [14] reported two cases of acute extremity ischemia due to BioGlue embolization. Mahmood and colleagues [15] recently described fatal myocardial infarction due to embolization of BioGlue into the coronary arteries after proximal aortic dissection repair.
We have also experienced a case of adhesive leak and possible embolization. A 68-year-old man with prosthetic aortic valve endocarditis and an ascending aortic aneurysm underwent reoperative aortic root replacement with a homograft. BioGlue was used to reinforce the suture lines. Intraoperative transesophageal echocardiography revealed a mobile lesion on the anterior leaflet of the mitral valve. A 5 x 10 mm fragment of BioGlue was removed from the mitral valve through a left atriotomy. The patient awoke with right hemiplegia and a computed tomographic scan revealed left middle cerebral artery thrombosis and acute stroke involving the frontal, temporal, and parietal lobes. This experience, along with the previous reports of adhesive embolization, prompted us to study the potential for BioGlue to leak through suture-line needle holes, as originally posited by Carrel and colleagues [12].
Complications related to surgical adhesives, including embolization, have been attributed to misapplication [21, 22]. The first 2 of Carrels 3 potential mechanisms of embolization can certainly be caused by technical error [12]. The third mechanism, entry of glue through suture-line needle holes, suggests that embolic complications can occur despite proper application in accordance with the manufacturers instructions. Our data demonstrate that adhesives can enter vessels and grafts through needle holes even when applied properly.
In the current study, leaks were significantly more common in fresh aorta than in prosthetic grafts. The very weak, friable nature of fresh aorta compared to the tough fibers of prosthetic grafts may account for this difference. Suturing the aorta often creates small linear tears, resulting in larger suture channels than those observed in prosthetic grafts. There was also a distinct difference in appearance of the leaks in the aorta compared with the leaks in prosthetic grafts. Prosthetic grafts had discreet, round particles, whereas the aortic tissue had thin, friable flakes that extended along the intimal surface. The characteristics of the different conduits may be responsible for these observations; the smooth, slick endothelium may encourage BioGlue to spread along the surface. The use of ePTFE suture, which is designed to improve needle hole sealing, did not eliminate adhesive leaks. Surprisingly, we observed no significant difference in leakage rates between various needle and suture sizes.
The finding that intraluminal adhesive particles were easily dislodged supports concerns regarding systemic embolization, a complication that may commonly go unrecognized. Given the significant morbidity associated with many major cardiac and vascular procedures, adverse events such as stroke and renal failure do not raise immediate suspicion of adhesive embolization. For example, Passage and colleagues [3] reported cerebrovascular accidents or transient neurologic deficits in 20 of 115 patients (17%) undergoing cardiac surgery with BioGlue. The authors, however, did not attribute the cerebral complications to BioGlue because it "binds solidly within seconds, does not crumble or fracture and therefore cannot embolize." Another reason the incidence of adhesive embolization may be underestimated is that postmortem microscopy examinations are often not obtained in patients who succumb to complications after cardiovascular procedures.
In 2001, BioGlue was approved in the United States [23] as an adjunct to standard methods of achieving hemostasis in adult patients undergoing open surgical repair of large vessels, specifically including the aorta, femoral, and carotid arteries. While the balance between BioGlues benefits and risks seems favorable in complex repairs, especially acute proximal aortic dissection, we caution against its routine use during cardiovascular operations. For example, BioGlue is marketed for use in carotid endarterectomies; the potential risk of BioGlue leakage and embolization, coupled with concerns regarding nerve toxicity, make its routine use in this setting worrisome [8]. We propose that the risks vs benefits for using BioGlue be considered on a patient-by-patient basis. BioGlue should be used only when necessary to optimize hemostasis and reinforce weak tissues.
When BioGlue use is warranted by the clinical situation, several technical adjustments may limit the risk of adhesive leakage through cardiovascular anastomoses. Whenever possible, we advise pressurizing the anastomosis before applying the glue. However, because BioGlue does not adhere to tissue in the setting of active bleeding, we apply the glue as reinforcement after achieving hemostasis with additional sutures; this reinforcing layer of BioGlue is meant to improve anastomotic strength and may be particularly beneficial in patients with extremely fragile tissue, such as those with Marfan syndrome. Slow release of the adhesive during application tends to increase its initial viscosity. Additionally, when applied to suture lines on the closed, clamped proximal aorta, temporarily stopping the left ventricular sump suction may reduce the chance of drawing glue into the lumen. Finally, whenever possible, direct inspection of the lumen after BioGlue application may allow detection of leaks.
As new uses for BioGlue in cardiovascular surgery are developed, additional technical modifications will need to evolve in order to avoid embolic complications. For example, several authors [3, 6, 14, 24] have reported using BioGlue during repair of intracardiac defects involving the aortic annulus, mitral annulus, and ventricular septum. A recent report by Fink and colleagues [6] is noteworthy in that the authors emphasized the importance of protecting the left heart chambers from contamination with the glue and described several technical precautions aimed at optimizing adherence and preventing embolization.
Finally, the limitations of our study warrant discussion. It is difficult to know how closely our simple model reflects the complex setting of intraoperative anastomoses. For example, the presence of blood within the field may alter the frequency and characteristics of adhesive leaks. This study does not address the wide variation in aortic fragility encountered in the clinical setting. We chose to use unpressurized anastomoses to demonstrate the worse case scenario. Notably, the only clinical trial [4] that has demonstrated an objective benefit with BioGlue required that the adhesive be applied to unpressurized cardiovascular anastomoses; ie, before clamp removal. Although additional studies will be required to address the relevance of these findings to the clinical arena, we have demonstrated that the potential for adhesive leakage exists and should be considered when making decisions regarding BioGlue use.
| Discussion |
|---|
|
|
|---|
When you presented this earlier data in Florida last year, I was very concerned what I might see. I have conservatively probably applied BioGlue now to probably about 1,000 anastomoses. After your presentation in Florida, I went back and I started looking with magnification, two and a half power, and I could never see any BioGlue droplets, and I looked at Dacron to tissue, Gore-Tex to Dacron, Gore-Tex to Gore-Tex, all sorts of different anastomoses. Again, this was in a clinical setting, and at least I dont think I have seen anything embolic that I have blamed on the glue. Your clinical cases obviously, as small as these droplets are, what you presented here, are probably not from needle hole leaks.
But I wonder if the fact that these anastomoses have blood on them, you know, which the BioGlue mixes with quickly, if maybe that is a reason why in a clinical setting that I am not seeing these droplets or anything inside, or maybe I am too old, maybe I am missing them?
DR LEMAIRE: One of the limitations of this study is that, rather than a field with blood and all the other things that go along with clinical anastomoses, these are just dry anastomoses; how we can apply this to the clinical setting is uncertain. This is a first step at looking at whether or not it is even feasible for this material to get through the suture line holes. I think we dont see a lot clinically because the particles are fairly small. Professor Carrels case illustrated this well, in that there was a patient who didnt have any clinical evidence of neurologic injury and yet had pieces of glue in the brain.
One of our concerns is that small amounts of embolization can certainly contribute to postoperative neurologic complications other than stroke, such as temporary neurologic dysfunction and neurocognitive deficits. All emboli are bad and adhesives may be adding to the emboli. So, despite the small size of these adhesive emboli, we remain concerned. But, like you, we havent seen any particles within our anastomoses. It is difficult to correlate what we have seen here in the lab with what you have described and what we have seen in the operating room.
DR ROBERT ROBBINS (Stanford, CA): I think he makes a very good point, and what I was going to suggest to you is maybe you want to go back and do this study again and then run blood through and pressurize and put a filter in and try to catch it at the end.
The second thing is the dissections that you presented, I have stopped putting it at the distal part of the anastomosis when you do an open distal and try to put glue and glue the distal, because I think it is getting into at least the false lumen and then through communications distally. I have not seen any cases to prove that, but I guess Craig must have. I wasnt aware of that.
And then finally, one question I wanted to ask you, have you looked at other sealants like CoSeal from Cohesion?
DR LEMAIRE: No, we havent. We havent had a lot of luck with the other sealants in terms of their ability to really strengthen some of these fragile aortic anastomoses, so we focused on the one that we use clinically. We do use BioGlue primarily for acute dissections and people with fragile aortic tissue like the Marfan patients. We were interested in seeing whether or not this specific agent would be a problem.
One of the other concerns that we had was that the viscosity of this glue is very low. It is quite runny, and we thought that may contribute to its ability to seep into the lumen. I know of some other cases that are going to be reported as case reports, even from our institution, where they have seen some problems with embolization, but these are macroscopic emboli that go along with the type of embolization after dissection repairs that you mentioned.
| Acknowledgments |
|---|
|
|
|---|
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
K. J. Zehr Reply. Ann. Thorac. Surg., September 1, 2008; 86(3): 1056 - 1057. [Full Text] [PDF] |
||||
![]() |
D. Spielvogel, M. N. Mathur, and R. B. Griepp Aneurysms of the Aortic Arch Card. Surg. Adult, January 1, 2008; 3(2008): 1251 - 1276. [Full Text] |
||||
![]() |
K. J. Zehr Use of Bovine Albumin-Glutaraldehyde Glue in Cardiovascular Surgery Ann. Thorac. Surg., September 1, 2007; 84(3): 1048 - 1052. [Abstract] [Full Text] [PDF] |
||||
![]() |
G. J. Van Nooten, P. Somers, R. Forsyth, K. Narine, Y. Van Belleghem, S. Jacobs, and F. De Somer Autologous glue: Part of the sticky mystery unraveled J. Thorac. Cardiovasc. Surg., August 1, 2007; 134(2): 415 - 423. [Abstract] [Full Text] [PDF] |
||||
![]() |
Z. Mitrev, V. Belostotskii, and N. Hristov Suture line reinforcement using suction-assisted bioglue application during surgery for acute aortic dissection Interactive CardioVascular and Thoracic Surgery, April 1, 2007; 6(2): 147 - 149. [Abstract] [Full Text] [PDF] |
||||
![]() |
P. Tansley, F. Al-Mulhim, E. Lim, G. Ladas, and P. Goldstraw A prospective, randomized, controlled trial of the effectiveness of BioGlue in treating alveolar air leaks J. Thorac. Cardiovasc. Surg., July 1, 2006; 132(1): 105 - 112. [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 |