Volume 8, Issue 1 , Pages 36-41, February 2003
Repair of spontaneous rupture of the posterior wall of the left ventricle after mitral valve replacement☆☆☆
Article Outline
One of the most serious complications in cardiac surgery is the occurrence of a rupture of the posterior wall of the left ventricle after mitral valve replacement. This is a challenging and often humbling experience to the cardiac surgeon, carrying a significant morbidity and mortality. The original reports of this complication were documented in the late 60s and early 70s.1, 2, 3 Since then, multiple reports of isolated cases or short series have appeared in the literature, describing the events and proposing different technical approaches to prevent and correct the problem. The etiologic factors have become well recognized, and preventive measures, including new developments and technical approaches to mitral valve surgery, have reduced markedly the incidence of this problem and virtually eliminated the potential occurrence of certain types of left ventricular rupture. Three specific types of rupture are described (Fig 1) and they have been classified according to their anatomic location.2, 4
Type I
A Type I rupture is located at the atrioventricular groove. This remains the most common site and can be seen in the following circumstances:
Type II
A Type II ventricular rupture occurs at the base of the papillary muscle, primarily due to excessive resection of the posterior papillary muscle, with local hemorrhage and rupture.
Type III
A Type III rupture is located between Type I and Type II lesions, and is most often related to posterior left ventricular wall trauma, due to a high profile or large prosthetic valves, often in combination with a small left ventricular cavity. With a bioprosthetic valve, injuries to the left ventricular wall are known to occur due to penetration of the valve struts in the posterior myocardium. This usually happens at the time of valve insertion or from elevating the heart from the pericardial cavity with a bioprosthesis in situ.
Ruptures have been documented after mitral valve replacement for mitral insufficiency following an acute myocardial infarction.6 These are ruptures that occur in less specific locations and, although they may be primarily related to myocardial necrosis, a combination with the above-described etiologic factors can trigger the event.
The evolution in techniques of mitral valve surgery over the past two decades has lowered significantly the occurrence of these complications and, in reality, Type II and III have been virtually abolished, primarily due to the following technical reasons:
Surgical technique
1 This figure illustrates the different anatomic sites of rupture of the posterior left ventricular wall. Type I ruptures, which occur in the atrioventricular groove, are the most common. The anatomic position of the coronary sinus and potentially of the circumflex coronary artery is demonstrated. This relationship must be kept in mind if injury to these structures during repair is to be avoided. With present techniques of mitral valve surgery, Type II and Type III ruptures are rarely observed and have been virtually abolished. An exception is the possibility of rupture at these sites following mitral valve replacement for ischemic mitral valve insufficiency, after acute inferior wall myocardial infarction. The presence of a prosthetic device may induce or contribute to rupture of the infarcted area.
2 Annular decalcification can be a difficult procedure, depending on the magnitude of the calcareous infiltration. If it is small and localized, removal becomes less traumatic and usually can be accomplished without problems. However, when the calcific infiltration is deep and extensive, and the typical calcified ring is present, the problem becomes more challenging. Not always we are able to extract the calcific ring completely and safely by a subendocardial dissection. An initial attempt should be made by incising the atrial endocardium close to the calcified ring; hopefully, we can find a dissection plane that would allow the surgeon to excise the calcium ring in its entirety (A). In our experience, this is very difficult to accomplish because of the depth and irregular infiltration of calcium in the myocardium. More often, the calcium is removed in small pieces with a forceps and rongeur. The utilization of an ultrasonic probe is reasonably safe and adds to a more thorough decalcification procedure. If a rupture should occur in this area, further decalcification may be necessary if appropriate repair of this rupture is to be accomplished. As shown in (B), the repair can be accomplished utilizing strips of Teflon on the ventricular and atrial sides, closing the gap by utilizing interrupted horizontal mattress sutures of 2-0 Prolene or similar sutures. Large gaps may require a patch, usually of autologous pericardium, sutured to the ventricular and atrial endocardium with running sutures of 3-0 Prolene, as illustrated in (C). These techniques are used initially during the mitral valve replacement surgery, or to repair a delayed Type I rupture. In this case, the prosthetic valve must be removed and re-implanted after repair is accomplished. The partially resected anterior leaflet and its attachments to the papillary muscle are sutured to the mitral annulus as per standard techniques.
3 The prosthetic valve is secured with interrupted 2-0 braided sutures. At the site of repair, these sutures will be placed through the strip of Teflon (A) or pericardial patch (B). The orientation of the sutures is a surgeon's preference and can be reversed, ie, going from the mitral annulus to the prosthesis. If a bioprosthesis is being used, the sutures holding the valve struts must be kept in place until the valve is completely seated in order to avoid potential penetration of the struts in the posterior wall.
4 Some patients undergoing mitral valve replacement for acute mitral valve insufficiency resulting from an acute transmural myocardial infarction can rupture the posterior wall after valve implantation. (A) Shows a cross-section view of the ventricles, the area of necrosis, and rupture after acute myocardial infarction. Limited lacerations can be repaired with full-thickness interrupted mattress sutures buttressed on Teflon felt pledgets by approximating the edges of the ruptured wall. This technique is applicable in small lacerations, but should be performed with the patient on cardiopulmonary bypass; otherwise, expansion of the lesion can occur, further complicating the repair. This repair is done externally, which is preferable to the internal repair because of the better exposure of the lesion and direct observation of the circumflex coronary artery anatomy. (B) Shows repair of a more extensive area of necrosis, which requires a patch to reconstruct the ventricular wall. The hematoma is resected and reconstruction of the posterior wall is accomplished with a pericardial or a fabric patch sutured to the edges of the defect with interrupted 2-0 polyester sutures (or similar), with Teflon felt pledgets on the endocardial side of the ventricle. As mentioned before, removal of the prosthetic device before attempting repair of such large lacerations increases the chances of a permanent and successful procedure. Also, after repair is accomplished, one may select a smaller prosthetic valve than originally used. We believe that if this approach is not used, then the repair would be less satisfactory and the incidence of recurrent rupture or false aneurysm would be greater. (C) The necrotic area is excised as close as possible to normal myocardium. The extent and site of rupture may preclude preservation of circumflex artery branches. (D, E) The patch is being attached to the ventricular muscle by the technique described. The patch must be properly configured and sized in order to assure proper left ventricular contour and, at the same time, avoid any tension on the suture line.
Comments
Aside from the standard techniques of repair of left ventricular posterior wall rupture herein described, different approaches have been utilized to deal with the problem, with a variable initial success rate.5, 6, 7, 8, 9 In some cases, the repairs are not long lasting, and early recurrence has been described, usually with catastrophic consequences. Late occurrence of false aneurysm post-repair has been reported.9, 10
To increase the odds of a successful repair, it is imperative that this is done with the patient on cardiopulmonary bypass, and after removal of the mitral valve prosthesis, although isolated cases of successful repair with the prosthesis in place have been reported. Some surgeons have gone to the extreme of performing this repair in a setting of autotransplantation.8 We believe this to be unnecessary. Proper exposure becomes vital for a successful repair, and a transseptal atrial approach should be used if necessary.
In cases of false aneurysm, the repair should be done as shown in Fig 2B and C. The prosthetic valve is removed and repair is accomplished from inside the heart, as depicted. The false aneurysm cavity should be left undisturbed in order to minimize the need for further manipulation of the heart.
One major issue is how to treat a hematoma in the posterior wall of the ventricle or atrioventricular goove after completion of mitral valve replacement. We have found this situation on two occasions in over a thousand mitral valve procedures. On both occasions, the hematoma was small and did not enlarge during a period of observation post-cardiopulmonary bypass, even with manipulation of the arterial pressure to observe potential consequences of moderate hypertension. They were left undisturbed and were inconsequential to these patients. Our concern with a possible delayed rupture or a false aneurysm formation did not materialize. We believe that a small, nonexpanding hematoma should be left alone and should be managed medically by preventing hypertension. On the other hand, a large and expanding hematoma must be dealt with, and should be resected and repaired as described. Indecision and prolonged observation will only compound the problem, making a successful surgical repair more difficult.
A Type I rupture is more conducive to internal repairs than is Type II or Type III. Because these repairs should be full thickness, enclosure of a circumflex branch in the repair may be unavoidable and may be inconsequential if the vessel is rather small and nondominant. However, if the vessel included is a branch of a dominant circumflex coronary artery, revascularization of the distal vessel should be performed, if possible. As stressed, attempts to elevate the heart from the pericardium for repair of the laceration, with the prosthetic valve in place, should be avoided. This would most likely expand the laceration, which further complicates an already difficult situation. If it must be done for whatever reason, the patient should be on cardiopulmonary bypass and the surgeon must use the pulmonary veins and not the atrioventricular groove as a fulcrum, elevating the entire heart and not only the ventricle from the pericardial cavity.
Ischemic muscle necrosis resulting from a major myocardial infarction is usually extensive, making the repair more difficult. However, cases of posterior wall left ventricular rupture after mitral valve replacement in patients after acute myocardial infarction have been described, and the majority occur as a Type III rupture.9 It is interesting to note that, in this series, all patients had mitral valve excision and replacement. Most probably, with present techniques of valve preservation, these events would not occur, or at least would be less frequent. Over the past several years, the great predominance of mitral valve repair and the preservation of leaflets and chordal attachments in cases of replacement have rendered this complication rare.
References
- . Causes of early postoperative death following cardiac valve replacement: clinicopathologic correlations in 64 patients studied at necropsy. J Thorac Cardiovasc Surg. 1967;54:422
- . Unusual complications during mitral valve replacement in presence of calcification of the annulus. Ann Thorac Surg. 1971;11:336
- Rupture of the posterior wall of the left ventricle after mitral valve replacement. J Thorac Cardiovasc Surg. 1975;69:259–263
- Intraoperative left ventricular rupture associated with mitral valve replacement. Chest. 1973;64:409
- . Reconstruction of the left ventricle with autologous pericardium. J Thorac Cardiovasc Surg. 1987;94:710–714
- Mitral valve replacement early after myocardial infarction: attendant high risk of left ventricular rupture. JACC. 1987;9:277–282
- . Successful repair of left ventricular rupture after redo mitral valve replacement. Ann Thorac Surg. 1989;47:916–917
- Autotransplantation of heart for repair of left ventricular rupture after mitral valve replacement. Transplant Proc. 2001;33:3553–3554
- Repair of left ventricular rupture after mitral valve replacement: use of a Teflon patch and glue. Ann Thorac Surg. 1999;67:1802–1803
- Pseudoaneurysm of the left ventricle following mitral valve replacement. J Cardiovasc Surg. 1990;31:242–246
☆ Address reprint requests to Anoar Zacharias, MD, FACS, FACC, 2213 Cherry St., ACC 309, Toledo, OH 43608.
☆☆ 1522-2942/03/0801-0000$35.00/0
PII: S1522-2942(03)70033-6
© 2003 Published by Elsevier Inc.
Volume 8, Issue 1 , Pages 36-41, February 2003
