Clinical Profile and Outcome in 52 Patients with Cardiac Pseudoaneurysm

Methods

Patients

The echocardiographic, surgical, and medical databases of the Mayo Clinic were searched for patients with pseudoaneurysm diagnosed between January 1980 and September 1996. The diagnosis was based on typical morphologic features shown by each imaging method, including echocardiography [23-32], angiography [33], computed tomography [34-36], and magnetic resonance imaging [23, 37]. Surgical or pathologic confirmation (or both) was available in 42 patients (81%). Although some authors have reported that pseudoaneurysm can occur as a complication of infectious endocarditis, especially of the aortic valve [6, 11], we decided not to include this subset of patients because of the difficulty of differentiating pseudoaneurysm from subannular abscess, even at surgery [7].

After identifying the study group, we reviewed the clinical records, which contained patient demographic characteristics, clinical presentation, results of surgical repair, and vital status at latest follow-up. Follow-up was obtained from the Mayo Clinic records and was supplemented by direct mail and telephone contact. Cause of death was determined on the basis of clinical records, death certificates, and correspondence.

Diagnosis of Pseudoaneurysm

The typical two-dimensional echocardiographic features of pseudoaneurysm [23-25] include a relatively narrow neck in comparison with the diameter of the aneurysm (Figure 1) and sharp discontinuity of the endocardium at the site at which the aneurysm communicates with the left ventricle. Features of the left ventricle on contrast angiography [33] include a narrow communication between the aneurysm and the left ventricular cavity and a paucity of coronary vessels in the vicinity of the pseudoaneurysm (Figure 2). Pseudoaneurysm was diagnosed on computed tomographic scans (Figure 3) if the myocardial wall ended abruptly at the border of the aneurysm [34-36]. Diagnosis made by magnetic resonance imaging was based on the presence of an aneurysm rimmed only by pericardium with a low signal [37].

Statistical Analysis

Continuous variables are expressed as means ±SDs. The small sample size prohibited informative statistical comparison between groups of patients.

Results

Demographic and Clinical Features

Using the standard imaging features described in the Methods section, we identified 52 patients with pseudoaneurysm: Thirty-four (65%) were male, and the mean age was 48 ± 28 years (range, 0.5 to 84 years). The pseudoaneurysm was discovered incidentally in 25 asymptomatic patients (48%). Cardiac imaging was performed in 13 of these 25 patients after staged repair of complex congenital heart disease, in 7 patients during follow-up after valve operations, and in 5 patients because of an abnormal electrocardiogram obtained while patients were at rest. Four patients (8%) presented acutely: 3 with acute myocardial infarction and 1 with cardiac tamponade. Other clinical presentations were congestive heart failure in 8 patients (15%), chest pain in 7 (13%), syncope or arrhythmia in 5 (10%), and systemic embolism in 3 (6%). Thirteen patients (25%) had a history of hypertension.

Diagnostic Methods and Location of Pseudoaneurysm

Pseudoaneurysms were imaged and diagnosed with two-dimensional Doppler echocardiography (38 patients), cardiac catheterization (33 patients), magnetic resonance imaging (5 patients), and computed tomography (4 patients). However, the diagnosis was initially made with two-dimensional echocardiography in 32 patients, cardiac catheterization in 12, magnetic resonance imaging in 4, and computed tomography in 2. In 2 other patients, the diagnosis was made intraoperatively during staged repair of tetralogy of Fallot and repeated aortic valve replacement.

The median interval between diagnosis and previous cardiac surgery or myocardial infarction was 9.1 months (range, 0 to 15 years) in all patients, 8.3 months in surgically treated patients, and 13.6 months in medically treated patients.

The location of pseudoaneurysms was related to their cause: After myocardial infarction, pseudoaneurysms were located in the inferior or posterolateral wall in 18 of 22 patients (82%); after surgery for complex congenital heart disease, they were located in the right ventricular outflow tract in 13 of 15 patients (87%); after mitral valve replacement, they were located in the posterior subannular region of the mitral valve in 4 of 4 patients (100%); and after aortic valve replacement, they were located in the subaortic region in 3 of 3 patients (100%).

Cause

As shown in Table 1, pseudoaneurysm in our series was caused by cardiac surgery (30 patients [58%]) and myocardial infarction (22 patients [42%]).

Follow-Up and Outcome

Complete follow-up to September 1996 was available for 48 patients (92%); the median duration of follow-up was 4 years (range, 3 days to 16.6 years). One patient was lost to follow-up after 14.8 months, 1 was lost after 4 years, 1 was lost after 11.8 years, and 1 was lost after 16.4 years.

Forty-two patients (81%) had either elective (38 patients [90%]) or emergent (4 patients [10%]) surgical repair (surgical group); the overall surgical mortality rate was 7% (3 deaths). Surgical or pathologic confirmation (or both) was available for all 42 patients who had surgical repair. The surgical mortality rate was higher among patients who had had myocardial infarction (13% [2 of 16] compared with 4% [1 of 26]).

Ten patients (19%) did not have surgery (medical group) for the following reasons: associated medical conditions (4 patients), small pseudoaneurysm (3 patients), and patient refusal to undergo surgery (3 patients). Six of these 10 patients had a pseudoaneurysm after myocardial infarction, and 4 had a pseudoaneurysm after cardiac surgery.

Nineteen patients died after a median survival of 2.3 years (range, 3 days to 8.2 years) after diagnosis of pseudoaneurysm. Of these 19 patients, 13 were from the surgical group; the causes of death were noncardiac in 5 patients, congestive heart failure in 3, myocardial infarction in 3, and documented ventricular tachycardia in 2.

Six medically treated patients died after a median survival of 2.1 years (range, 11 days to 4.7 years) after diagnosis: Three died of noncardiac causes, 2 died of congestive heart failure, and 1 died of acute myocardial infarction. No further cardiac rupture was documented. The overall survival rate at 2 years was 63%.

Table 2 shows the clinical characteristics and outcomes of patients in the two treatment groups.

Discussion

Cardiac rupture is usually fatal and accounts for 7% to 10% of early deaths after acute myocardial infarction [3-5]. In most cases, the cardiac wall ruptures into the pericardial cavity and causes cardiac tamponade and death. Contained rupture of the heart is recognized less frequently than cardiac rupture, and its incidence is unknown. Contained rupture has been attributed to pericardial adhesions in the area of rupture or a slow extracardiac leak that results in pericardial inflammation and adhesions [8]. Thus, an expanding intrapericardial hematoma, or pseudoaneurysm, is formed. The wall of the pseudoaneurysm consists of fibrous tissue and pericardium and lacks any component of the cardiac wall. Various reports suggest that such a contained rupture has a greater propensity for rupture than a true aneurysm, whose wall contains myocardium [2, 6, 19]. Information on the clinical course of pseudoaneurysms is gleaned from many reports of small series [1, 2, 6-40], and no large clinical series from a single institution has been reported. Acute rupture, surgical survival, and exceptional survival are overrepresented in patients with pseudoaneurysm. To our knowledge, our report describes the largest clinical series of patients with pseudoaneurysm from a single institution.

Clinical Presentation

The diagnosis of pseudoaneurysm is rarely suggested by clinical signs and symptoms. The patients in our series usually presented with nonspecific signs and symptoms. The lesion was found incidentally in 25 patients (48%). Patients who had pseudoaneurysm after myocardial infarction tended to present with recurrent anginal chest pain or congestive heart failure; this is consistent with descriptions in previous reports [8, 24, 38]. Other presentations included arrhythmia or syncope (5 patients) and systemic embolism (3 patients).

Cause

Cardiac pseudoaneurysm can occur after myocardial infarction, cardiac surgery, infectious endocarditis, and trauma to the chest [1, 2, 6-40]. Surgical procedures frequently associated with pseudoaneurysm include mitral valve replacement [6, 7, 9], aortic valve replacement [12, 14], and surgical correction of congenital heart disease [13]. These pseudoaneurysms tend to occur in areas of surgical manipulation or the sites of incisions, such as the site of a ventriculotomy or vent placement [15, 16]. In our series, an incisional pseudoaneurysm developed in two patients after repair of a true aneurysm and in three others after uncomplicated open heart surgery. Mitral valve replacement and aortic valve replacement have been associated with pseudoaneurysms. After mitral valve replacement, pseudoaneurysms may develop at sites of inadvertent periannular trauma; they are characteristically found in the posterior subannular region. It is postulated that removal of excess leaflet and annular tissues, selection of an excessively large prosthesis, suturing technique, and previous mitral valve replacement predispose patients to pseudoaneurysm after mitral valve replacement [7, 9].

After aortic valve replacement, pseudoaneurysms commonly occur in the subaortic region [41]. The postulated mechanisms for this include postoperative hyperkinesis that causes abnormal pressures within a cul-de-sac; inherent weakness of the aortic root; and weakness induced at surgery by tying sutures too tightly, decalcifying too vigorously, and using too many closely placed sutures. Surgery for congenital heart disease, particularly correction of tetralogy of Fallot and right ventricular outflow tract reconstruction or conduit implantation, have been associated with pseudoaneurysm [13].

The pseudoaneurysms that occurred after myocardial infarction were typically diagnosed within 6 months after the infarction, although intervals as long as 12 years have been reported [8]. In our series, the median time to diagnosis after myocardial infarction was 3.9 months; 55% of patients presented within the first 6 months. These pseudoaneurysms were usually located in the inferior or posterolateral walls, a finding consistent with previous reports [7, 8]. The reason for this predominance is unclear. In comparison, fatal cardiac ruptures tend to occur more often in the anterior and lateral left ventricular walls. Thus, anterior-wall rupture tends to be characterized by a more acute hemodynamic effect and catastrophic outcome, whereas rupture of the inferior or posterior wall tends to be contained; this difference explains the observed incidences [42].

Other, less common causes of pseudoaneurysm include trauma [16], infectious endocarditis [11], syphilis [43], rheumatic myocarditis [44], and disseminated tuberculosis [45]. Post-traumatic pseudoaneurysms are typically found on the anterior wall of the left ventricle [17, 39], and pseudoaneurysms occurring after infectious endocarditis tend to be located in the mitral-aortic intervalvular fibrosa [11].

Diagnosis

Various imaging methods have been used to diagnose pseudoaneurysm, including two-dimensional echocardiography, computed tomography, magnetic resonance imaging, and left ventricular angiography [23-31, 33-37]. Chest radiography sometimes shows a localized bulge on the cardiac silhouette [10, 40]. This feature was present in only five of our patients and has been reported by others [25].

Contrast angiography was the first imaging method used to detect pseudoaneurysms. The presence of a periventricular sac that communicated with the left ventricle through a narrow orifice was considered most suggestive of a pseudoaneurysm. This finding differs from findings in a true aneurysm, which typically has a large communicating neck. In addition, pseudoaneurysms appear avascular on coronary angiography [33].

The echocardiographic detection of pseudoaneurysm was first reported in 1975 [32]. The echocardiographic features typical of pseudoaneurysms include sharp discontinuity of the endocardial image at the site of communication of the pseudoaneurysm with the left ventricular cavity and an orifice that is relatively narrow in comparison with the diameter of the pseudoaneurysm [24]. Pulsed-wave and color flow Doppler echocardiography have been useful in visualizing the high-velocity, turbulent, bi-directional flow between the left ventricle and a pseudoaneurysm [26-31].

On computed tomography, pseudoaneurysms are characterized by an abrupt disappearance of the myocardial wall at the border of the pseudoaneurysm [34-36]. Magnetic resonance imaging shows the low signal of the pericardium, which constitutes the only wall of the pseudoaneurysm [37].

All of these imaging techniques are suitable diagnostic techniques short of surgical or pathologic confirmation. Each has its advantages and disadvantages. Other associations, such as valvular regurgitation, thrombus formation, and ventricular function, are often important additions to clinical management. This explains why echocardiography has become the most common examination used for first recognition.

Outcome

In our study, the surgical mortality rate was 7% for the 42 patients who had surgical correction. This rate is better than that reported previously [6] and is ascribed to improved surgical techniques and, possibly, patient selection. Previous studies generally advocated early surgery [2, 6-810, 19, 22, 23] because of the possibility of spontaneous rupture. Primarily on the basis of case reports, medically treated patients were assumed to have a poor prognosis, with cardiac rupture as the likely cause of death [2, 6]. In a review of 35 articles published between 1984 and 1993 (66 patients) on the management of pseudoaneurysm after myocardial infarction, Natarajan and colleagues [42] concluded that medical treatment of chronic pseudoaneurysm (>3 months) was not associated with increased risk for cardiac rupture. Sakai and colleagues [9] showed that pseudoaneurysm after mitral valve replacement (n = 8) may not always be associated with a poor prognosis. In their study, 7 of 8 patients who received conservative treatment had no complications.

For various reasons, 10 patients (19%) (6 in whom pseudoaneurysm developed after myocardial infarction and 4 in whom it developed after cardiac surgery) in our series did not have surgery for pseudoaneurysm. These patients provide insight into the natural history of pseudoaneurysm. Most of them (80%) had chronic pseudoaneurysm (>3 months), and no patient died as a direct result of pseudoaneurysm. Death was most strongly related to the primary underlying disease and cardiac dysfunction, which would not have been expected to have changed with or without repair of the pseudoaneurysm. Thus, the natural history of cardiac pseudoaneurysm may not be as catastrophic as suspected, but it should be noted that only a small number of patients were treated medically. Furthermore, our retrospective review cannot account for factors that may have guided the initial therapeutic decision.

With the widespread availability of routine noninvasive imaging, it is likely that more unsuspected pseudoaneurysms will be detected. This raises a question about optimal treatment, especially if a pseudoaneurysm is detected incidentally in an asymptomatic patient. On the basis of our experience, we recommend surgical resection as the treatment of choice and believe that it should be performed in patients with ventricular tachycardia or recurrent embolism related to the pseudoaneurysm; patients with other indications for cardiac surgery, such as coronary artery bypass surgery or staged repair of congenital heart disease; and patients in whom the pseudoaneurysm is detected within 3 months after myocardial infarction [42]. In asymptomatic patients who have high surgical risks because of associated medical problems, conservative management may be considered. Such patients can be monitored with noninvasive examinations.

The diagnosis of pseudoaneurysm in our patients was based on the accepted morphologic features seen with imaging techniques. Pseudoaneurysm was confirmed surgically in 42 patients and pathologically in 3 patients. Our study was not intended to scrutinize or compare definitive or absolute pathologic features but rather to determine the outcome of patients who had the accepted imaging criteria for pseudoaneurysm. In addition, it was impossible to exclude inherent patient selection bias, exclude false-positive results, or confirm the incidence of pseudoaneurysm.

The cause of death was based on clinical records, death certificates, and correspondence. We classified death as “cardiac” if it was due to congestive heart failure, myocardial infarction, documented arrhythmia, or a thromboembolic event. Of note, no documented acute events were consistent with late aneurysm rupture in either the surgically treated or medically treated group. Outcome was related most strongly to the underlying medical and cardiovascular status of the patient.

The clinical profile and long-term outcome of 52 patients with cardiac pseudoaneurysm discovered before death revealed that 1) except for the underlying condition, pseudoaneurysm had no characteristic clinical sign; 2) surgical repair of pseudoaneurysm was associated with an acceptable surgical mortality rate [7%]; 3) cardiac rupture did not occur in either surgically or conservatively treated patients; and 4) late death was related primarily to the underlying disease or cardiac dysfunction.

From Mayo Clinic and Mayo Foundation, Rochester, Minnesota; and Mayo Clinic Jacksonville, Jacksonville, Florida.

Dr. Malouf: Mayo Clinic Jacksonville, 4500 San Pablo Road, Jacksonville, FL 32224.

Drs. Oh and Seward: Mayo Clinic, 200 First Street SW, Rochester, MN 55905.