|
|
||||||||
Ann Thorac Surg 2007;83:526-531
© 2007 The Society of Thoracic Surgeons
a Departments of Thoracic and Cardiovascular Surgery, Cleveland Clinic, Cleveland, Ohio
b Quantitative Health Sciences, Cleveland Clinic, Cleveland, Ohio
Accepted for publication June 30, 2006.
* Address correspondence to Dr Navia, Cleveland Clinic, Department of Thoracic and Cardiovascular Surgery, 9500 Euclid Ave, Desk F24, Cleveland, OH44195 (Email: naviaj{at}ccf.org).
| Abstract |
|---|
|
|
|---|
METHODS: From January 1986 through December 2001, 30 patients aged 50 to 85 years (mean, 68; 70% male) underwent left ventricular pseudoaneurysm repair. Two surgical approaches were used: primary repair (n = 5, 17%) and patch closure (n = 25, 83%). Twenty-one patients (70%) had concomitant procedures, including coronary revascularization (n = 17, 57%) and mitral valve surgery (n = 9, 30%); 8 patients (29%) underwent emergent surgery. Clinical presentation, preoperative imaging data, and surgical outcomes were abstracted from medical records or obtained by patient follow-up.
RESULTS: The most common clinical presentations were heart failure (n = 22, 73%) and angina (n = 11, 41%). Pseudoaneurysm was rarely suspected at clinical presentation. Contrast ventriculography was diagnostic in 54% of patients in whom it was performed, as opposed to 97% for two-dimensional echocardiography (p = 0.2). Postoperative intra-aortic balloon pump was required in 7 patients (23%). Hospital mortality was 20%, and late survival was 73%, 59%, and 45% at 1, 5, and 8 years, respectively.
CONCLUSIONS: Left ventricular pseudoaneurysm should be suspected in postinfarction patients with unexplained heart failure. Echocardiography is usually diagnostic and is superior to ventriculography. The surgical mortality rate is elevated in this complex patient population. Long-term survival is also poor, mainly because of underlying ischemic cardiomyopathy.
| Introduction |
|---|
|
|
|---|
Except for case reports [4, 5] and small case series, few studies have addressed this rare postinfarction complication. Knowledge of the clinical profile and surgical results is sparse because of limited surgical experience in any single institution [6]. Further, the natural history of postinfarction ventricular pseudoaneurysm is not well understood. A high risk of expansion and rupture seems true of large pseudoaneurysms [7]. Thus, early surgical intervention is recommended, but operative risk is substantial because of underlying ischemic cardiomyopathy. Long-term survival after surgical treatment has yet to be determined.
Thus, the purposes of this study were to (1) describe the clinical presentation of LV pseudoaneurysm, (2) compare two-dimensional echocardiography and contrast ventriculography as diagnostic tests, and (3) determine operative and late surgical results.
| Patients and Methods |
|---|
|
|
|---|
Patients with postsurgical pseudoaneurysm or true postinfarction ventricular aneurysm were excluded. All patients suffered acute MI, confirmed by typical electrocardiographic changes and elevation of cardiac enzymes. At the time of MI, 8 patients (27%) were treated with thrombolytics. The most common location of MI was the posterior wall (n = 11, 39%), followed by the lateral wall (n = 8, 29%), inferior wall (n = 6, 21%), and anterior wall (n = 3, 11%).
Preoperative patient characteristics are listed in Table 1. Twenty-one patients (70%) were male, and mean age was 68 ± 8 years (range, 51 to 81).
|
Surgical Technique
Once the diagnosis of ventricular pseudoaneurysm was made, surgical treatment proceeded. Eight patients (28%) underwent emergent operation because of signs of cardiovascular collapse or imminent cardiac rupture (cardiac tamponade).
Surgical approach depended on whether the patient had a previous cardiac operation, pseudoaneurysm on the anterior wall of the left ventricle or near the undersurface of the sternum, cardiovascular collapse or cardiac tamponade requiring emergent operation, or associated cardiac problems requiring surgical treatment.
Median sternotomy with hypothermic cardiopulmonary bypass (CPB) was used in all patients. Although femoral vessels were exposed in patients who had undergone previous heart surgery, this site was used for instituting CPB in only 4 patients (3 electively because of cardiac tamponade or close proximity to the sternum, and 1 emergently because of patent graft injury after chest reentry). In all other patients, ascending aorta and double venous or double-stage venous cannulation was performed. Antegrade and retrograde cold crystalloid or cold blood cardioplegia was delivered every 15 minutes after the aorta was clamped.
With the anatomy identified, especially location and extent of the MI, the decision about surgical technique was made. Surgical procedures performed are listed in Table 2.
|
Concomitant procedures were performed in 21 patients (70%; see Table 2). The most frequent was coronary artery bypass grafting. Mitral valve repair and replacement were equally distributed, with 4 patients undergoing each procedure. Repair techniques were anuloplasty with a Cosgrove-Edwards band (Edwards Lifesciences, Irvine, California; n = 2), edge-to-edge leaflet repair [8] and anuloplasty with a Cosgrove-Edwards band (n = 1), and resuspension of the posteromedial papillary muscle (n = 1). All patients who underwent mitral valve replacement received a Carpentier-Edwards porcine or bovine prosthesis, size 27 or 29 mm. Four patients had inferior ventricular septal defects in association with ventricular pseudoaneurysm. The repair technique was similar to the one described by David and colleagues [9]. Mean CPB and aortic clamp times were 165 ± 76 minutes and 99 ± 35 minutes, respectively.
Follow-Up
Follow-up was obtained in part through consultations for echocardiographic assessment at Cleveland Clinic as well as through reports from referring cardiologists. Additionally, all patients or their family members were cross-sectionally followed using an Institutional Review Boardapproved protocol, which waived the need for patient consent. Mode of death was assigned through review of clinical records and death certificates. Follow-up was obtained in all but 1 international patient (97%). Mean follow-up among survivors was 52 ± 37 months (range, 0.5 to 130).
Data Analysis
Descriptive data are presented as means and standard deviations for continuous variables (medians with 15th and 85th percentiles for variables with skewed distributions) and frequencies and percentages for categorical variables.
Survival analysis
A parametric method was used to resolve the number of phases of instantaneous risk of death and to estimate shaping parameters [10]. For multivariable analyses, noninformative means imputation was employed for sporadic missing values to ensure that cases were not excluded when many variables were under consideration.
Bootstrap bagging was employed for variable selection in the hazard function domain (variables examined are listed in the Appendix) [11]. Random sampling with replacement was used to create 500 data sets the same size as the original, and then automated forward stepwise variable selection was used with p values of 0.07 for variable entry and 0.05 for retention in the model. Frequency of occurrence of variables in these 500 models was compiled; appearance of a variable in at least 50% of models was indicative of reliability.
Because of small sample size and only 13 deaths, we determined that the final model should contain only one, or at most two, risk factors. One risk factor was statistically significant at p less than 0.02 and was confirmed in more than 50% of the bootstrapped samples.
| Results |
|---|
|
|
|---|
Diagnostic Testing
Diagnostic investigation of ventricular pseudoaneurysm included contrast ventriculography in 24 patients (83%), transesophageal echocardiography in 23 (79%), transthoracic echocardiography in 21 (72%), and magnetic resonance imaging (MRI) in 2 (7%).
On contrast ventriculography, a paraventricular mass with a narrow neck and paucity of coronary vessels in the vicinity was typically seen. On cross-sectional echocardiography, morphologies were similar to those observed during cardiac catheterization, and diagnosis was aided by a sharp discontinuity of the endocardium at the site where the pseudoaneurysm communicated with the left ventricle. In addition, there was continuous Doppler flow signal extending from the LV cavity to the aneurysmal cavity through a narrow neck. On MRI, the presence of an aneurysm (detected by low signal strength) rimmed only by pericardium suggested the diagnosis of pseudoaneurysm.
Contrast ventriculography was diagnostic in 54% of the 24 patients in whom it was performed, as opposed to 97% among 23 patients in whom two-dimensional echocardiography was performed. Magnetic resonance imaging was diagnostic in the 2 patients in whom it was performed.
Surgical Outcomes
In-hospital outcomes
In-hospital surgical outcomes are listed in Table 3. Six patients died in-hospital. Modes of death were cardiogenic shock in 3, sepsis in 2, and uncontrolled bleeding from surgical site in 1. Six patients (20%) required reexploration for bleeding. Postoperative intra-aortic balloon pumping was required in 7 patients (23%) with cardiogenic shock, despite optimum inotropic support. Eight (27%) had prolonged mechanical ventilation requiring tracheostomy, and 4 (13%) had acute renal failure requiring dialysis. Median intensive care unit and postoperative hospital stays were 2 days and 10 days, respectively (see Table 3).
|
|
|
| Comment |
|---|
|
|
|---|
Diagnostic testing
Definitive diagnosis usually involves more than one test. At the beginning of our series, contrast ventriculography was the gold standard, demonstrating the characteristic narrow mouth of an aneurysm compared with the width of the fundus [15]. Now, echocardiography, either transthoracic or transesophageal, is the diagnostic standard [16]. Besides demonstrating myocardial discontinuity [17] and turbulent flow characteristics within the chamber [18], it also provides important information about anatomy, size and location of the defect, associated valvar disease, and estimate of ventricular function. Transesophageal echocardiography seems to provide more accurate information than transthoracic echocardiography, particularly of a posterior pseudoaneurysm [19]. Although not statistically significantly different in our experience, echocardiography was superior to contrast ventriculography, with a diagnostic accuracy of 97%. Because so few of our patients had MRI, we were unable to compare it with other diagnostic tools. However, MRI is reported to be highly accurate in determining size and location of the pseudoaneurysm [17]. Additional advantages include the ability to distinguish between pericardium, thrombus, and myocardium and the potential to visualize disruption of the epicardial fat layer by the pseudoaneurysm. The primary limitation of MRI is its lack of portability.
Surgical outcome
Our surgical mortality is similar to that reported in the literature. Lower body mass index and greater degree of LV dysfunction were identified as risk factors for death. In addition to poor LV function [13], mitral valve operation, reoperation, and operation during the acute phase of MI [20] have been identified as risk factors for surgical mortality. Although not observed by us, recurrent rupture has been reported [21] and was nearly always fatal. Five-year survival in our study was 59%, similar to that of other series [22]. Most of our late deaths were from ischemic cardiomyopathy, emphasizing the importance of coronary revascularization at the time of pseudoaneurysm repair. Recurrent LV false aneurysm has been reported [23]. Mechanisms include extension of infarction adjacent to the site of a true aneurysm, reinfarction distal to an occluded coronary artery bypass graft, progression of coronary arteriosclerosis, early and late breakdown of primary closure, and suture dehiscence due to infection. Mechanism of recurrence in our experience was patch dehiscence (2 patients).
Natural History and Indication for Operation
Because of its rarity, the natural history of ventricular LV pseudoaneurysm is unknown. Rupture with fatal exsanguination is not inevitable [24]; however, the pseudoaneurysm has little strength and is prone to rupture. Although rupture of a large false aneurysm is unpredictable and can occur at any time [24], the natural history of small ones is unknown; some authors believe the risks are similar [15].
The most frequent location of a false ventricular aneurysm is the posterior wall [6, 12], because anterior or lateral LV ruptures are unlikely to be contained, resulting in sudden death. More than half our patients had undergone previous cardiac operation, which can explain the occurrence of pseudoaneurysm in every other location. Nevertheless, if containment of rupture is protective on the one hand, it increases the complexity of operation and surgical risk on the other.
We believe that surgery is indicated for everyone as soon as the diagnosis is established, unless the surgical risk is prohibitive [6], especially for large or expanding pseudoaneurysms, regardless of symptoms. This approach is justified by uncertain natural history in regard to rupture and risk of embolizing thrombotic material. Small chronic pseudoaneurysms that are diagnosed incidentally, which constituted 6% of our cases, seem to have a more stable course. Pretre and colleagues [20] suggest, however, that chronic and asymptomatic pseudoaneurysms less than 3 cm in diameter and without evidence of expansion might be followed conservatively [20].
Surgical Treatment
Surgical treatment is complex. Acute rupture must be considered, especially when the pericardium is opened. We strongly recommend that the surgical team be prepared to institute CPB quickly. Pseudoaneurysms of the anterior wall or in proximity to the undersurface of the sternum may pose particular dangers. Before opening the sternum, it is wise to cannulate femoral vessels, establish CPB, and institute hypothermic circulatory arrest [25]. Femorofemoral bypass before opening the chest has been advocated for hemodynamically unstable patients and for those in cardiac arrest, because external cardiac massage is contraindicated [21]. When opening the pericardium or during reoperations, only cannulation sites should be freed from adhesions, because of the risk of rupture and dislodgment of intramural thrombus. Manipulation and dissection of the left side of the heart should be performed only after the aorta is clamped and the myocardium protected.
In all our patients, including those who had not undergone previous cardiac operation, dense and vascular adhesions were found. This finding supports the notion that an intense inflammatory reaction is necessary to contain the rupture, which would otherwise be fatal.
Various techniques have been used to obliterate the pseudoaneurysm neck [20]. Our choice of technique was based on extent of the defect and geometric relation between the defect and papillary muscles, and status of the myocardium. Usually by 4 weeks after MI, mature hyalinized fibrous tissue tough enough to anchor sutures has formed, and primary repair is warranted [26]; that was performed in 17% of our patients. Use of sutures placed through strips of PTFE felt is advisable to avoid bleeding from the suture line or from needle holes in the myocardium. The majority of our patients had large ventricular pseudoaneurysms or ones located close to the basal part of the LV. In these cases, reconstructing the LV with a patch is preferable, to avoid distortion of the mitral valve apparatus [26] or excessive traction on the ventricular edges [20]. Treating ischemic mitral regurgitation (30% of our patients) and ventricular septal defect (13%) is also important.
Limitations
This is a single-institution clinical cohort study of a small number of patients. Additionally, prevalence of ventricular pseudoaneurysm after MI was not possible to estimate (the "denominator") because only surgical cases were included.
In conclusion, LV pseudoaneurysm should be suspected in post-MI patients with unexplained heart failure. Echocardiography is superior to contrast ventriculography as a diagnostic test. Prompt and carefully planned surgical treatment is required to also address underlying coronary artery disease and other mechanical complications of MI. Surgical repair is challenging and early mortality is elevated in this highly complex patient population. Long-term survival is also poor, mainly because of underlying ischemic cardiomyopathy.
| Appendix |
|---|
Symptoms: New York Heart Association class (IIV), Canadian Cardiovascular Society angina class (04), interval from myocardial infarction to surgery
Ventricular Function: Left ventricular function (catheterization), left ventricular function (echocardiogram), evidence of myocardial infarction on electrocardiogram, left ventricular ejection fraction
Valve Pathology: Mitral valve regurgitation (grades 04)
Cardiac Comorbidities: Any cardiac operation, number of previous cardiac operations, angina, heart failure
Laboratory Values: Creatinine (mg · dL1), bilirubin (mg · dL1), hematocrit (%)
Coronary Anatomy: Number of systems diseased (
50%); three-system disease (
50%); maximum left main trunk stenosis; maximum left anterior descending coronary artery (LAD) stenosis, LAD system disease 70% or greater; left circumflex (LCx) disease (%), LCx disease 50% or greater, LCx disease 70% or greater, any LCx disease
Procedure: Concomitant coronary artery bypass grafting surgery
Experience: Date of operation, expressed as years since January 1, 1986.
| Acknowledgments |
|---|
|
|
|---|
| Footnotes |
|---|
|
|
|---|
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
I. E. Konstantinov, P. Saxena, S. Ling, and T. Barrett Repair of acute postinfarction atrioventricular dehiscence in a patient with posterior left ventricular aneurysm. J. Thorac. Cardiovasc. Surg., January 1, 2009; 137(1): 247 - 249. [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 |