| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Ann Thorac Surg 2001;71:1894-1899
© 2001 The Society of Thoracic Surgeons
a Mississippi Baptist Medical Center, University of Mississippi Medical Center, Jackson, Mississippi, USA
b Department of Surgery, University of Mississippi Medical Center, Jackson, Mississippi, USA
Address reprint requests to Dr McMullan, Mississippi Baptist Medical Center, 501 Marshall St, Suite 100, Jackson, MS 39202
e-mail: cvsc{at}ucmail.com
Presented at the Forty-seventh Annual Meeting of the Southern Thoracic Surgical Association, Marco Island, FL, Nov 9–11, 2000.
 |
Abstract |
|---|
 Top Abstract IntroductionMaterial and methodsResultsCommentAddendumDiscussionReferences |
|---|
Methods. Since 1980, we have operated on 18 patients with LVFWR. Fourteen patients had experienced "blow-out" rupture associated with cardiogenic shock. Four patients had "stuttering" ruptures, a less spectacular occurrence. Echocardiography was the most important diagnostic tool. Repair was performed, usually using infarctectomy and direct suture closure.
Results. Eleven patients (61%) died after operation, 4 patients as a result of rerupture 1 to 12 hours after operation. Recently, we have used a "patch/glue" technique to repair ruptures in 2 patients. We believe this technique is superior to direct suture closure in preventing rerupture. There have been 7 long-term survivors (39%) from 6 months to 15 years.
Conclusions. Left ventricular free wall rupture is not always sudden and dramatic. Yet, the operating staff must be willing to race to the operating room even with the patient in full resuscitation. Echocardiography is the most sensitive and efficient diagnostic tool. All rupture sites should be aggressively repaired, possibly combining direct suture and patch/glue techniques.
|
Introduction |
|---|
|
TopAbstractIntroductionMaterial and methodsResultsCommentAddendumDiscussionReferences |
|---|
We published the surgical results of LVFWR in our first 4 patients in 1984 [5]. This series has now expanded to 18 patients undergoing operative management of this often-dramatic malady. We have chosen to divide our patients somewhat arbitrarily into two groups according to clinical presentation. Patients with "blow-out" rupture are those experiencing a period of cardiogenic shock (blood pressure less than 70 mm Hg for at least 10 minutes) before surgical repair with or without resuscitation. "Stuttering" rupture is used to describe patients with symptoms or signs that wax and wane but do not create significant hemodynamic instability. This group is characterized clinically and during operation by a relatively small leak that may seal spontaneously, only to leak again.
|
Material and methods |
|---|
|
TopAbstractIntroductionMaterial and methodsResultsCommentAddendumDiscussionReferences |
|---|
|
Four patients had stuttering ruptures, 2 of whom were referred by a single outlying hospital and transported 90 miles after echocardiographic diagnosis before undergoing successful repair.
Surgical technique
All patients were repaired with the aid of cardiopulmonary bypass. Femorofemoral cannulation was used in 13 hemodynamically unstable patients whereas 5 patients had standard aortic and right atrial cannulation. Intermittent aortic cross-clamping was used in the first 5 patients, cold blood cardioplegia was used thereafter. Repair was usually performed (first 16 patients) with direct suture closure over felt strips with or without preceding infarctectomy. After a brief period of aortic cross-clamping, during which time the rupture site was identified and opened more widely to decompress the heart, the infarct margins were palpated between two fingers with the heart beating. This maneuver helped the surgeon decide on the need for infarctectomy (7 patients) in an especially large, necrotic infarct. The heart was fibrillated in the first 5 patients, and the remaining patients underwent cardioplegia to enhance closure. Smaller infarcts were closed directly. Three patients underwent concomitant coronary artery grafting.
The most recent 2 patients represent our experience with the patch/glue technique. Due to its availability we used standard off-the-shelf Instant Krazy Glue (Elmer’s Products Inc, Columbus, OH), a cyanoacrylate product, in combination with a glutaraldehyde-treated bovine pericardial patch (Baxter Healthcare Corp, Edwards CVS Division, Irving, CA) to cover and seal the rupture. The glue is sterile by virtue of its composition but the tube container is not. The glue is spread liberally straight from its tube over the entire appropriate size patch, carefully maintaining sterile conditions. This must be done rapidly to maximize glue function and requires a relatively dry surface for application, which can be accomplished on cardiopulmonary bypass. The first attempt was in a patient with posterior rupture. The second patient with left main coronary artery occlusion ruptured during the operation upon opening the pericardium. In addition to patch/glue repair of the rupture, this patient underwent coronary artery grafting (left anterior descending and circumflex).
Six patients underwent insertion of the intraaortic balloon assist device during operation to promote discontinuation of cardiopulmonary bypass. No patient had intraaortic balloon assist insertion before the operation.
|
Results |
|---|
|
TopAbstractIntroductionMaterial and methodsResultsCommentAddendumDiscussionReferences |
|---|
The fourth patient with suspected rupture underwent cardiac catheterization when echocardiography showed only a small to moderate pericardial effusion. He was found to have occluded the right and left anterior descending coronary arteries with 90% stenosis of the circumflex. Our intention was to explore for rupture and bypass the right, diagonal, and circumflex coronary arteries. During the operation, dark, bloody, pericardial fluid was present but under no pressure. There was diffuse anterior and apical subepicardial hemorrhage with no obvious rupture site. After coronary grafting he came off cardiopulmonary bypass without difficulty. Eleven hours and 40 minutes later, he suddenly developed profuse bleeding into the chest bottles with electromechanical dissociation. It took 17 minutes to return to the operating room and establish femorofemoral bypass. The anterior apical rupture was successfully repaired using infarctectomy and direct suture closure using felt strips. However, he died 3 days later of brain death when the family elected to discontinue the ventilator.
In the 14 patients with blow-out rupture, 4 died during operation of cardiac failure. One of these had successful repair but after decannulation protamine infusion was initiated and resulted in an anaphylactic reaction. (She had undergone two previous aortocoronary bypass procedures.) Despite reinstitution of cardiopulmonary bypass this patient could not be resuscitated.
Six patients were weaned from cardiopulmonary bypass and in 5 bleeding was controlled. One patient experienced bleeding from extension of the rupture/repair site into the atrioventricular groove and died 12 hours later of uncontrolled hemorrhage.
The rupture site could not be identified in 1 patient in this group early in our experience. The infarct was posterior near the atrioventricular groove as viewed at operation. There was a large hematoma in the atrioventricular groove, and we did not explore it for fear of disruption. He came off cardiopulmonary bypass satisfactorily, but after 15 minutes in the cardiovascular recovery unit (CVR) he experienced rerupture. After the patient was raced to the operating room, we established femorofemoral bypass in 7 minutes. The rupture site was approximately 1 cm to the left of the ventricular septum and 1 cm from the atrioventricular groove. We were able to excise a small portion of the infarct and direct suture close over felt strips. Again, the patient came off bypass but died 24 hours later, unresponsive and anuric.
Two patients reruptured in the CVR, both 12 hours after successful repair. Efforts to repeat the repair in operation were unsuccessful in 1 patient. In the other the chest was opened in the CVR but resuscitation was unsuccessful.
One patient died of progressive renal failure and brain death 10 days after operation. An 86-year-old man experienced a left brain CVA 3 days after successful repair and died of respiratory failure 17 days after operation. Causes of death in nonsurvivors are listed in Table 2.
|
|
|
Comment |
|---|
|
TopAbstractIntroductionMaterial and methodsResultsCommentAddendumDiscussionReferences |
|---|
Certain clinical features may yield clues to the diagnosis. Age, sex, hypertension, diabetes mellitus, use of thrombolytic agent, and location of the infarct are not helpful because of variations from report to report. There may be a predilection to first infarctions and even single-vessel disease because of the paucity of collateral vessels [6]. Some series have reported a high incidence of positional pleuritic chest pain, unprovoked emesis without preceding nausea, and unexplained restlessness or agitation [7].
The electrocardiogram may show unrelenting or progressive ST segment elevation after the infarct [8] On physical examination there may be signs of tamponade to include pericardial friction rub, quiet heart, paradoxical pulse, or elevated jugular venous pulsation [9]. The single most accurate diagnostic sign is electromechanical dissociation with a 97.6% predictive accuracy [10].
In any postinfarction patient with sudden deterioration, one must have a high index of suspicion and consider the diagnosis of LVFWR. An echocardiogram, the sine qua non of diagnosis, should be obtained immediately. In this setting the presence of a substantial pericardial effusion should precipitate rapid mobilization of the surgical team.
In the event of progression to cardiogenic shock or cardiorespiratory arrest, we believe a race to the operating theater to establish femorofemoral bypass is essential to provide the best chance for survival. Meanwhile, inotropic support is invaluable in resuscitation, and we use external cardiac massage if necessary, although its effectiveness is questionable with a blow-out rupture and tamponade. Pericardiocentesis, intraaortic balloon counter pulsation, and even bedside emergency cardiopulmonary support require valuable time that may sacrifice organ function, especially the brain.
Once femorofemoral bypass is established, rapid sternotomy should be performed. If upon opening the pericardium the rupture site is obvious, one may or may not conclude that some degree of infarctectomy is necessary. It may be helpful to attempt approximation of ruptured or resected edges with suture backed by felt strips. Most importantly, use of cyanoacrylate glue (Instant Krazy Glue) placed rapidly on a large patch of glutaraldehyde-treated bovine pericardium and applied over the rupture site and the surrounding infarct, appears to be the most effective method in controlling bleeding and in preventing rerupture. Four of our patients experienced rerupture as the cause of death. With the present patch/glue technique, these patients would be potentially salvageable, as would the patient who had undergone two previous aortocoronary bypass procedures who died of protamine anaphylaxis. With this last patient, we waited 30 minutes after discontinuing cardiopulmonary bypass before decannulating and giving protamine because of our previous experience with rerupture. Unfortunately, the need for recannulation appears to have created critical ischemia time leading to this patient’s death. As a result of this experience, we now give protamine with the cannulas in place to any patient at increased risk for protamine sensitivity, such as previous exposure to protamine or protamine zinc insulin, shellfish allergy, or previous vasectomy.
Although this decision is controversial, we believe cardiac catheterization is contraindicated if LVFWR is suspected. In patients with stuttering rupture, the argument could be made for preoperative catheterization to define the coronary anatomy. However, it is difficult to know if such a rupture will allow enough time to accomplish this objective safely. In this time of aggressive catheter intervention it seems wise to consider cardiac catheterization and percutaneous transluminal coronary angioplasty/stenting at a later date if symptoms, nuclear scanning, or coronary anatomy dictate.
Patients with clinical evidence of LVFWR, with or without echocardiographic confirmation, in whom the rupture cannot be identified at operation can present a dilemma. We have found it possible to identify the site of rupture by briefly cross-clamping the aorta with the heart beating after going on cardiopulmonary bypass. By increasing the intracavitary left ventricular pressure, rerupture will occur. Alternately with the patch/glue technique, it is possible to cover the infarct area completely and be confident that the rupture is contained.
In the remarkable series reported by Padro and colleagues [11], 13 patients with "subacute" rupture were successfully treated with the patch/glue technique, yielding 13 long-term survivors. We would probably have classified most of their patients as having stuttering ruptures in view of their utilizing preoperative hemodynamic studies using a Swan–Ganz catheter at the bedside in addition to echocardiography. Nevertheless, it is commendable to obtain 100% survival in this critically ill group of patients.
The survival rate in our series of frequently desperately ill patients with LVFWR was 39% (7 of 18). However, with the patch/glue technique our 4 patients with rerupture represent potentially salvageable patients. Furthermore, the patient with protamine anaphylaxis presents another possible rescue situation using current techniques. These 5 patients added to the actual survivors yield a potential survival rate of 67% (12 of 18) in this series.
In conclusion, LVFWR is not always sudden and dramatic. Echocardiography is the most sensitive and efficient diagnostic tool. One must be willing to race to the operating room even with the patient undergoing full resuscitation. All rupture sites should be aggressively repaired, possibly combining direct suture and patch/glue techniques. However, some patients cannot be salvaged because of the inability to manage large blow-out ruptures with tamponade and still preserve vital organ function.
|
Addendum |
|---|
|
TopAbstractIntroductionMaterial and methodsResultsCommentAddendumDiscussionReferences |
|---|
The first patient is a 73-year-old woman who presented to our hospital with an acute inferior myocardial infarction and congestive heart failure. Cardiac catheterization revealed severe three-vessel coronary arteriosclerosis and severe mitral regurgitation. On December 26, 2000, she underwent attempted mitral valve repair, subsequent mitral valve replacement, and three-vessel coronary artery bypass. At termination of cardiopulmonary bypass she was noted to have a small (less than 1 cm) LVFWR at the lateral edge of the inferior infarct. This rupture was repaired by direct suture backed with felt. The intraaortic balloon pump was required to separate the patient from cardiopulmonary bypass. Twelve hours later she reruptured, requiring emergency opening of the sternotomy wound in the cardiovascular intensive care unit and a rapid transit to the operating room. Repair the second time was with the suture/felt pledget technique initially to achieve a dry surface and then by covering the entire area of infarction and rupture with the patch/glue technique. She has weathered postoperative coagulopathy, emergency reopening of the sternotomy in intensive care unit because of tamponade, secondary closure of the sternum 24 hours later, respiratory failure requiring tracheostomy, and renal failure requiring dialysis. Currently she remains hospitalized on dialysis but is continually improving. It appears she is likely to become a long-term survivor as her renal function improves.
The second patient is a 55-year-old man who presented to our emergency room with an acute inferior myocardial infarction. He was treated with lytic therapy and heparin. Within 12 hours he developed a loud systolic murmur at the left sternal border, which on echocardiogram was found to represent a large midseptal, posterior VSD. The following day catheterization revealed 99% right coronary stenosis and 60% stenosis of the midleft anterior descending. QpQs was calculated at 4:1. Left ventricular free wall rupture was not suspected. Operation on February 6, 2001, revealed gross blood under modest pressure within the pericardium. In addition to a prominent thrill over the right ventricle, there was a large hematoma covering the entire posterior wall with the most prominent hematoma located in the area of the atrioventricular groove. This finding represented a "stuttering" rupture that had sealed. The VSD was approached through a linear incision in the midposterior wall just to the left and parallel to the ventricular septum. The VSD was a "Swiss cheese" like defect 3 to 4 cm in diameter. It was closed with large mattress polypropylene sutures placing felt pledgets on the left ventricular posterior surface of the VSD and passing the sutures through the ventricular septum to the external free wall of the right ventricle. The incision in the left ventricle was closed with polypropylene over felt strips. The entire posterior wall from the atrial side of the atrioventricular groove to the apex of the left ventricle was covered using the patch/glue technique. The left anterior descending was bypassed using a saphenous vein graft. No bypass was inserted into the right coronary system. Intraaortic balloon pumping was required to terminate cardiopulmonary bypass, but 3 days later the balloon pump was removed. Currently the patient is being ventilated and is on dialysis. He has a residual VSD with a calculated QpQs of 1.6. His long-term survival remains problematic.
|
Discussion |
|---|
|
TopAbstractIntroductionMaterial and methodsResultsCommentAddendumDiscussionReferences |
|---|
If infarctectomy is to be performed, do you now use a free wall patch sutured into the defect or do you still simply pull the edges together? It would seem to me that Professor Jatene’s principles about establishing the geometry of the ventricle with a ventricular patch would apply here just as in ventricular aneurysm operations.
Number two, how do you handle the difficult problem of the rupture dissecting through the ventricular wall? In your study you describe a case in which the rupture dissected all the way to the atrioventricular groove. If you apply the patch to the epicardium without infarctectomy, is then dissection more likely to occur?
You frequently mention the "race to the operating room" in these patients in cardiogenic shock. Might you not establish femorofemoral extracorporeal membrane oxygenation in the intensive care unit and then walk to the operating room with a stable patient?
Finally I wondered about the potential role of the use of left ventricular assist devices in these patients. One might be able to better wean patients from cardiopulmonary bypass and provide some degree of left ventricular decompression, albeit through left atrial cannulation, but that might prevent early rerupture of the ventricle.
Your presentation was excellent and we are grateful to you for your contributions.
DR McMULLAN: Generally the way we have handled infarctectomy is to apply the aortic cross-clamp after going on bypass while the heart is beating, then open the ventricle to decompress it. By palpating the wall between two fingers, one can get a good idea of how much of the myocardium is involved. If it is a large area of myocardium with extensive necrosis, we have in the past resected the area of muscle involved and then closed by pulling the two edges together. More recently, though, with the patients who have undergone the patch/glue technique, we believe we can preserve geometry much better by using the patch/glue technique and not resect any muscle.
Dissecting rupture, especially to involve the atrioventricular groove, is a difficult problem. In our early experience we failed to recognize the importance of repairing all ruptures. In the particular patient with rerupture described in our report, we returned to the operating room and were able to perform infarctectomy staying far enough away from the atrioventricular groove that we did not get into trouble. Unforunately, though, he died as a result of brain hypoxia. We believe the patch/glue technique offers the opportunity to contain a dissecting rupture without radically altering geometry.
As far as walking or racing to the operating room with femorofemoral bypass, we went through a period of time during which we had an emergency cardiopulmonary support cart available for use in such a situation. Unfortunately, emergency cardiopulmonary support also requires time to use and is difficult for us to implement with external massage and other resuscitative measures taking place. We have found that, at least in our institution, it is better for us to get to the operating room as rapidly as possible and then make use of femorofemoral bypass.
Your comments about left ventricular assist devices are interesting and certainly represent something to take into consideration in the management of these patients in the future.
DR ROBERT GUYTON (Atlanta, GA): I appreciate the opportunity to discuss this nice report. I would rise to endorse the use of the glue technique and endorse the technique of not resecting muscle, and I think you emphasized that in your answer to Dr Pennington. Currently we would not, except in very unusual circumstances, attempt to resect the myocardium and would use the patch glue technique.
One adjunct that I think is important that you used in several of your patients is a balloon pump. I think the balloon pump does decrease the left ventricular peak pressure in the postoperative period. We tend to use the balloon pump for 72 hours postoperatively, whether the patient needed it hemodynamically or not, to try to prevent rerupture; in your operative survivors you had about a 30% rerupture rate. Can you comment on what you would do in the perioperative period to prevent rerupture and whether or not you think either ventricular assist or the balloon pump has a rule in this situation. Thank you.
DR McMULLAN: First, let me thank the discussants. Doctor Pennington was my esteemed classmate in medical school and Dr Guyton is the son of my esteemed medical school physiology professor.
As to Dr Guyton’s comments, to prevent rerupture and control bleeding the patch/glue technique offers considerable advantage. We did have 6 patients who required the intraaortic balloon assist device in the postoperative period, none of whom experienced rerupture. But, again, I think the most important factor to prevent rerupture is use of the patch/glue technique.
Thank you.
|
References |
|---|
|
TopAbstractIntroductionMaterial and methodsResultsCommentAddendumDiscussionReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  A. D'Onofrio, N. Abbiate, P. Magagna, and A. Fabbri Intraoperative coronary angiography in postinfarction ventricular free wall rupture: how technology can change diagnostic and therapeutic timing Interactive CardioVascular and Thoracic Surgery, August 1, 2008; 7(4): 733 - 735. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. W.M. van den Borne, J. Narula, J. W. Voncken, P. M. Lijnen, H. T.M. Vervoort-Peters, V. E.H. Dahlmans, J. F.M. Smits, M. J.A.P. Daemen, and W. M. Blankesteijn Defective intercellular adhesion complex in myocardium predisposes to infarct rupture in humans. J. Am. Coll. Cardiol., June 3, 2008; 51(22): 2184 - 2192. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. F. Padera Jr. and F. J. Schoen Pathology of Cardiac Surgery Card. Surg. Adult, January 1, 2008; 3(2008): 111 - 178. [Full Text]
|
|
|
|
 |
|
 F. A. Atik, J. L. Navia, P. R. Vega, G. V. Gonzalez-Stawinski, J. M. Alster, A. M. Gillinov, L. G. Svensson, B.G. Pettersson, B. W. Lytle, and E. H. Blackstone Surgical Treatment of Postinfarction Left Ventricular Pseudoaneurysm Ann. Thorac. Surg., February 1, 2007; 83(2): 526 - 531. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G. Weissman, C. C. Kwon, R. K. Shaw, and J. F. Setaro Free-Wall Rupture of the Myocardium Following Infarction: A Changing Clinical Portrait in the Reperfusion Era: A Case Report Angiology, October 1, 2006; 57(5): 636 - 642. [Abstract] [PDF]
|
|
|
|
 |
|
 X.-M. Gao, Q. Xu, H. Kiriazis, A. M. Dart, and X.-J. Du Mouse model of post-infarct ventricular rupture: time course, strain- and gender-dependency, tensile strength, and histopathology Cardiovasc Res, February 1, 2005; 65(2): 469 - 477. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Fukushima, J. Kobayashi, O. Tagusari, and Y. Sasako A huge pseudoaneurysm of the left ventricle after simple gluing of an oozing-type postinfarction rupture Interactive CardioVascular and Thoracic Surgery, March 1, 2003; 2(1): 94 - 96. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
F. J. Schoen and R. F. Padera Jr. Cardiac Surgical Pathology Card. Surg. Adult, January 1, 2003; 2(2003): 119 - 185. [Full Text]
|
|
|
|
|
|
P. Totaro and G. Minzioni Patch-and-glue technique for left ventricular free wall rupture Ann. Thorac. Surg., July 1, 2002; 74(1): 294 - 295. [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 |
