|
|
||||||||
Ann Thorac Surg 2000;70:993-998
© 2000 The Society of Thoracic Surgeons
a Department of Medicine, University of Mississippi School of Medicine, Jackson, Mississippi, USA
Address reprint requests to Dr Bennett, Division of Cardiovascular Diseases, University of Mississippi Medical Center, 2500 North State St, Jackson, MS 39216
e-mail: kbennett{at}medicine.umsmed.edu
| Abstract |
|---|
|
|
|---|
| Introduction |
|---|
|
|
|---|
In 1599, the anatomist Realdus Columbus described how the necropsy of a young man, who had failed to recover from a syncopal episode, revealed the absence of the pericardium. Although some anatomists credited him with the original description of pericardial absence, others doubted the accuracy of his observation, suggesting he mistook an adherent parietal pericardium for an absent one [cited by Moore, 1]. In 1793, Matthew Baillie, MD, physician to St. Georges Hospital, London, and renowned for his diagnostic skill using palpation, reported the case of a 40-year-old man whose pericardium was totally absent. He described the finding with such clarity that there can be no doubt as to what he had seen [2]. Inquiries concerning the mans history and clinical findings provided little information. Baillie, not having examined the subject during life, and being unable to find that any particular observations had been made on the pulse by the attending physicians, therefore concluded:
It is not fair to conclude from thence that the action of the heart is exactly the same whether it be inclosed or not in its proper membrane; but it is surely enough to shew that the action is not very much affected by the want of it.
In 1925, Moore reviewed defects of the pericardium, and, including the accounts of Columbus and Baillie, collected 64 examples discovered at autopsy. He commented on the characteristic paucity of symptoms, the frequent finding of pleuropericardial adhesions, and the overwhelming tendency for the defect to occur on the left side [1].
In their 1938 review [3], Southworth and Stevenson examined the accounts cited by Moore and any published afterwards. After tracing each to the original source, excluding doubtful cases and those associated with gross fetal malformations, they settled on 45 as true examples. And they added one of their own, the only case found in a series of over 14,000 autopsies done at the Johns Hopkins Hospital. With one exception, the defects were on the left and, in 35 (76%), there was a common serous cavity containing the heart and left lung. In the remaining 11 (24%), there was only a foramen (partial defect) with the heart remaining inside the pericardium, although in 3 it was "protruding," but in only 1 of the 3 did it seem the probable cause of death. A woman, age 28, 3 days postpartum, collapsed after a fit of "urgent dyspnea." When her chest was opened, the heart was found in the left pleural cavity where it had lodged after escaping through a 3/4 inch sickle shaped defect in the left pericardium. Her death, reported by Boxall in 1887, is likely the first documented death from incarceration [4]. The mean age at death (excluding infants under 1 year) was 43.9 years, about that expected for the population at the time. Males outnumbered females three to one. About half had unexplained heart enlargement, and death in 25% was associated with "pleuropericarditis." They concluded that pericardial absence or incomplete development was compatible with a normal life, but individuals so affected did have a significant morbidity risk from exposure of the heart to lung infections.
Subsequent additions to the literature bolster the view that although there is obviously a loss of the hearts barrier to infection, the mere presence of the condition does not threaten life, for it is the size and location of the defect that are critical. If the left parietal pericardium is so underdeveloped that the heart and lung share the left pleural space, the heart can move freely without risk of entrapment, and the outlook is generally favorable. This form of pericardial deficiency accounts for about 75%, and the clinical, radiographic, and electrocardiographic features have been well described [5]. On the other hand, if the heart is enclosed except for a foramen that is so located that herniation can occur, then death may follow incarceration of all or a part of it [4,610]. This form of pericardial deficiency accounts for only 25% of the defects and is the subject of this review.
| Case report |
|---|
|
|
|---|
|
|
Two years later, she had a second episode of pneumonia and was noted to have an elevated left hemi-diaphragm. She had no more chest pain and, until stopped by joint discomfort, she could walk 2 to 3 miles without any problem. Ten years after surgery, palpitation was bothersome to the point that atenolol was begun. When last examined, at age 51, 20 years after surgery, her blood pressure was 180/95 mmHg, and there was paralysis of the left hemi-diaphragm. An electrocardiogram (ECG) showed a nonspecific T wave abnormality. An echocardiogram showed paradoxical interventricular septal motion, but was otherwise unremarkable.
| Material and methods |
|---|
|
|
|---|
| Results |
|---|
|
|
|---|
Fatal cases
In none (of the 5 cases) was the fatal attack preceded by symptoms. In two accounts, the victims, boys aged 2 and 4 years, were playing when the attack began [6,8], and in one instance, a 4-year-old boy had been fighting with his sister [9]. With the remaining 2, there was no exertion or unusual activity before the episode [7,10]. Unlike the death reported by Boxall [4] in which the entire heart had herniated into the left chest, in these cases there was incomplete escape with cicumferential strangulation of the body or apex of the left ventricle, and a deep groove of the myocardium separated the normal and incarcerated parts.
Clinical presentation of nonfatal cases
In one-third (13 of 39), an abnormality (hilar prominence or "bulge") on a chest film made for reasons unrelated to the defect per se (eg, respiratory infection, heart murmur, routine physical examination) initiated investigations leading to the diagnosis [11, 1214 (case 2), 15, 16 (case 2), 1719 (case 1), 2023]. In the rest [14 (case 1), 16 (case 3), 19 (case 2), 2445], that is, those with a complaint likely due to the defect, seemingly innocent symptoms varied from palpitations to chest discomforts variously described as sharp, stabbing, fleeting, or shooting. More disturbing complaints included chest pressure or constriction, sweating, dyspnea, circulatory collapse, and syncope. Symptoms were usually unrelated to any particular activity, but exertional angina pectoris, angina-like discomfort [14, 26, 35, 37], and myocardial infarction [26, 38, 39] did occur. In 3, pericarditis was the cause of pain [25, 29, 41]. There was a single account of chest discomfort with recurring pleural effusion [31]. Syncope occurred in 5 [16, 19, 28, 36, 45], and marked hypotension bordering on circulatory collapse occured in a 28-year-old man with an incarcerated atrial appendage [21].
In 12, the physical examination was abnormal, notably, for example, for a single patient with pectus excavatum [45], 2 with a pericardial rub [29, 41], 5 with a left sternal edge-pulmonic ejection murmur [11 (case 1), 16, 21, 27, 43], 2 with a systolic mitral murmur [14, 25], and 2 with an apical or left sternal edge diastolic murmur [32, 43].
Chest radiograph
In every instance, the heart was in the normal position, with varying prominence of the left heart border, due to herniation of a frequently enlarged left atrial appendage. This in turn was attributed, sometimes erroneously, in order of increasing frequency to the following: ventricular aneurysm, mitral valve disease, tumor of the hilum or mediastinum, or idiopathic dilatation of the pulmonary artery. In the one case in which the hilum appeared normal, the defect was confined to the apex of the ventricle and the heart appeared spherical with lung interposed between it and the diaphragm [39].
Electrocardiogram
Of the 13 cases in which the diagnosis was made only after discovery of a chest film abnormality, the ECG was reported in 12. Eleven of the tracings were normal but, in one instance, a "nonspecific ST-T abnormality" was noted. In the rest, that is, the 26 with symptoms that prompted investigation, the electrocardiogram was typically normal unless the defects rim impinged the left ventricle. In 11 cases in which surgery showed that the rim of the defect extended on to the left ventricle, an ECG was obtained in 9, and was abnormal in all but 3. An abnormality (ST-T) of repolarization was present in 4, and anterior infarction in 2 [38, 39].
Diagnosis
In earlier articles, the diagnostic methods were either angiocardiography (levophase of a pulmonary angiogram) showing the atrial appendage moving beyond its expected confines; or demonstration of air or carbon dioxide within the pericardial space following an artificial pneumothorax. After 1975, pneumothorax has only rarely been used. Today, two-dimensional echocardiography, contrast-enhanced computer assisted tomography (CT), transesophageal echocardiography, and magnetic resonance imaging (MRI) are most often employed [22, 34, 39]. Finally, thoracoscopy has been used to confirm the diagnosis after failure of CT and MRI to do so [38].
Surgery
Thirty-four patients had surgery and, in all but 4, the diagnosis was suspected beforehand [12, 2426]. Two were explored for a mediastinal tumor [12, 24], and 1 for a mistaken diagnosis of coronary atherosclerosis [26]. With six exceptions [23, 26, 37, 38, 41, 44], the surgeon used a left thoracotomy incision. Risher [38], Vanderheyden [41], Rusk [44] and their colleagues resected pericardium using the thoracoscopic technique, and Yamagishi and associates [23] closed a 3 cm defect by the same method. Operations involved amputation of the appendage with [13, 21, 28, 29, 45], or without [14, 43] defect closure; closure without amputation [15, 17, 19, 23, 24, 30, 3436]; and no amputation, but with a pericardiotomy or pericardiectomy [16, 2527, 3133, 3739, 41, 42, 44]. In 3, the defect and appendage were left as found [12, 14 (case 2),20]. In 2, a myocardial revascularization procedure was done [26, 37]. Defect closure, other than primary suturing, involved the use of a pleural flap [17, 21], autologous pericardium [45], synthetic fabric patch [29, 35], or xenograft [34, 36].
Findings at surgery
The defects were round to oval, ranging in size from 1 x 1 cm to 8 x 8 cm. Typically, the protruding left atrial appendage was larger than normal, often elongated, and freely movable within the defect. The largest appendage was 11 cm long x 2.5 to 3.5 cm wide, resembling a "giant dog ear" [13]. Acute and chronic inflammation, thickening, adhesions, and scarring, were often described. In only two instances was the appendage strangulated [21, 29]. The rim of the defect was usually thickened, rolled, and in the 11 instances in which the defect also encompassed part of the left ventricle, it typically indented the ventricle and compressed or constricted the coronary vessels. In one, the rim was calcified [42]. Constrictive calcific pericarditis was found in one [25]. The course of the left phrenic nerve varied with the location and size of the foramen, and tended to run along the antero-medial free rim of the defect [8, 14, 15, 17, 19, 24, 29, 35]; but, in two instances, it ran behind it [33, 45]. Kaneko and associates [49] described a "split" phrenic nerve with ventral and dorsal segments straddling the defect.
| Comment |
|---|
|
|
|---|
Radiographic features
With a defect limited to the base of the heart, the silhouette should be normal except for a prominent bulge along the left border produced by the herniated appendage. Yet this signal abnormality which reflects absence of pericardial reflection over the base of the heart, sometimes underestimates the danger. An apical defect on the other hand, which is more likely to be lethal, might not distort the heart silhouette at all. If the defect encompasses more than the base of the heart, exposing a portion of the ventricle, the atrial appendage invariably distorts the hilum, and the taut rim may constrict the left ventricle sufficiently to produce a recognizable groove. (Fig 2).
Electrocardiogram
The ECG is typically normal, but an abnormality of repolarization (ST-T) suggestive of coronary insufficiency, implies compression of the coronary circulation by the rim of the defect.
Differential diagnosis and confirmation
Presumption of the diagnosis requires only observing a structure (eg, appendage, pulmonary artery) outside its normal confines, but confirmation requires demonstrating the actual absence of pericardium. Such demonstration is necessary to eliminate conditions with a similar appearance, in which the pericardium is intact, eg, an enlarged pulmonary artery, mitral disease with enlargement of the appendage, tumor, pericardial cyst, or the rare aneurysm of the appendage [43, 46, 47].
Angiocardiography is limited in that actual demonstration of the presence or absence of pericardium is beyond the limits of the method. Artificial pneumothorax is well suited for demonstrating absence of pericardium, but not as reliable in defining the anatomy and extent of the defect. Because of discomfort and morbidity, it is now rarely used, yet there remains the occasional situation in which it can be helpful [40].
Of the noninvasive procedures (transthoracic and transesophageal echocardiography, CT with contrast, and MRI), MRI is probably the best way to show both absence of pericardium and the defects relationship to the underlying structures [22, 39]. However, since imaging at the level of the atria and ventricles requires epicardial fat or pericardial fluid for tissue differentiation, a paucity of fat such as may occur in some individuals, particularly children, may hinder definition of the pericardium. Even though the extent of the defect may be uncertain, the isolated protrusion of the left atrial appendage, coupled with indentation of the ventricle by the defects rim is very likely, as Gassner and his associates [39] suggest, "pathognomonic." Thoracoscopy may be the best of all confirmatory methods, since it is possible to remedy the problem without an additional procedure [23, 38, 41, 44].
Etiology
The cause, if anything more than an unfortunate accident of development is unknown, and there is little evidence for a hereditary pattern, although one family has been reported in which a 9-year-old boy and his half-sister both completely lacked a pericardium [48]. It is widely held that the disorder represents persistence of the embryonic pleuro-pericardial foramen [1, 6], perhaps due to inadequate blood supply following premature atrophy of the left common cardinal vein (duct of Cuvier), an event which could account for both the defect and its left-sidedness [1]. A recent article suggests that some defects may be due to a tear in the pleuro-pericardial membrane rather than failure of the pleuro-pericardial foramen to close [49].
Associated disorders
Judging from the accounts comprising this review, the defect usually occurs as an isolated disorder. Nevertheless, defects have been found during surgery for a variety of cardiothoracic disorders, eg, tetralogy of Fallot, atrial and ventricular septal defect, patent ductus arteriosus, single ventricle, cor triatriatum [11, 50], and bronchopulmonary malformations [51].
Management
If the defect circumscribes part of the left ventricle, especially the body or apex, surgery is clearly warranted, even if the coronary circulation is not compromised. On the other hand, if the defect is confined to the upper left heart border, permitting only the appendage to protrude, the condition is not particular hazardous, for although incarceration occurs [21], it rarely does. Unlike the rare intrapericardial aneurysm of the left atrial appendage [43, 46, 47], which it so closely resembles, the partial pericardial defect is not associated with thromboembolic events or atrial arrhythmias; therefore, if the individual is symptom free, there is no compelling reason for surgery.
Enlarging the defect may be all that is needed, and in some cases this is possible using a thoracoscopic approach [23, 38, 41, 44]. A decision to revascularize is problematic, for even with a generous pericardiectomy, coronary artery narrowing may remain. If the coronary lesion appears "fixed," it will probably persist [26, 27, 37], but "migratory" narrowing will likely correct [38]. In any event, closing the foramen to restore the hearts natural barrier to infection is appealing, therefore, an operative technique that permits defect closure is theoretically preferable. Whatever the approach, the surgeon should be mindful of the left phrenic nerve, which although typically runs along the antero-medial rim of the defect, occasionally will be found behind [33, 45], or even straddling it [49].
Concluding remarks
The pericardium may not be "vital," but it is far more than a simple sac containing fluid to provide a relatively frictionless heart action. It is not only a barrier to infection, but it maintains ventricular interdependence and restrains expansion of the hearts chambers during periods of volume or pressure overload. Furthermore, the fluid contains small amounts of complement, other immune factors, prostaglandins and myocardial cellular enzymes, implying a biological role far beyond its service as a lubricant [5254].
More than 20 years have elapsed, however, since the woman whose account is here reported underwent near total resection of the pericardium; apart from paralysis of the left hemi-diaphragm resulting from phrenic nerve injury, and altered ventricular septal motion, she has been little affected. Thus, it appears, as Baillie concluded over two centuries ago [2], that "the action [of the heart] is not much affected by the want of it."
Although uncommon, congenital partial absence of the left parietal pericardium is a potentially grave condition, one that can result in pleuropericarditis, angina pectoris, myocardial infarction, or even death. The following conclusions are offered: Firstly, the absence of symptoms, even in patents with a normal ECG does not always portend a good outcome. Secondly, if the defect is associated with chest discomfort of any description, and the ECG is abnormal, the likelihood for the potentially lethal defect is great. Finally, if symptoms are present, surgery is probably warranted regardless of defect location.
| Acknowledgments |
|---|
|
|
|---|
| Footnotes |
|---|
|
|
|---|
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
J. F. Bruzzi, M. Remy-Jardin, D. Delhaye, A. Teisseire, C. Khalil, and J. Remy When, Why, and How to Examine the Heart During Thoracic CT: Part 2, Clinical Applications Am. J. Roentgenol., February 1, 2006; 186(2): 333 - 341. [Abstract] [Full Text] [PDF] |
||||
![]() |
H M Tee and A T Carty A heart problem? Br. J. Radiol., March 1, 2003; 76(903): 211 - 212. [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 |