(Circulation. 1996;93:2197-2202.)
© 1996 American Heart Association, Inc.
Articles |
Pleural and Pericardial Effusions in a 50-Year-Old Woman
Presented March 13, 1995, at the University of Texas Health Science Center Medical School at Houston.
From the Department of Internal Medicine (E.B., R.H., P.F.W.) and the Department of Pathology and Laboratory Medicine (K.A.H.), University of Texas Health Science Center at Houston.
Correspondence to Eddy Barasch, MD, University of Texas Health Science Center at Houston, MSB 1.257, 6431 Fannin, Houston, TX 77030.
Key Words: cardiac tamponade • Clinicopathological Conferences • echocardiography • pericarditis
 |
Case Presentation (Ramesh Hariharan, MD) |
|---|
 Top Case Presentation (Ramesh... Hemodynamic Measurements During...Clinical Discussion (Eddy...Pathological Findings (Kent A....Clinical Follow-upFinal DiagnosisReferences |
|---|
A previously healthy 50-year-old Chinese woman presented with a low-grade fever, a generalized headache, and chest pain of 2 to 3 weeks' duration. She described intermittent sharp chest pain over the precordium that intensified when she lay down and shortness of breath after minimal exertion. Despite having lost her appetite, she had gained
10 pounds during the past month
and had noticed ankle swelling. She denied having had arthralgia or
skin rash and had not experienced any nocturnal dyspnea, wheezing,
cough, expectoration, or hemoptysis. She had visited her relatives in
the Middle East 6 months earlier, but her past medical history was
uneventful. She had taken acetaminophen (Tylenol) tablets
and a Chinese herbal preparation, but her symptoms continued. On physical examination, she appeared weak and ill. Her temperature was 36.7°C (98.0°F), and her pulse was 110 beats per minute, regular, and had normal volume and character. Her blood pressure was 115/70 mm Hg, which decreased to 90/70 mm Hg on inspiration; her respiratory rate was 22 breaths per minute. Carotid pulsations were normal. Jugular veins were distended to the angle of the mandible when the patient sat upright, but no further venous engorgement was noted on inspiration. There was mild mucosal pallor, but the oropharynx was otherwise normal. The first and second heart sounds were normal, and there were no clicks or gallops. A superficial scratchy systolic sound was heard intermittently over the left lower sternal region. Dullness to percussion, scattered inspiratory crackles, and diminished air entry were evident over both lung bases. Abdominal examination demonstrated a soft, tender liver palpable 2 cm below the right costal margin. Her pelvis and rectum showed no abnormality. A stool guaiac test was negative. Neurological examination was normal. There was moderate pitting edema below the level of the knees.
Results of the initial laboratory studies appear in Tables 1
and 2, and the results of chest
roentgenography, electrocardiography, and
echocardiography are presented in the
"Clinical Discussion." A CT scan of the chest showed bilateral
pleural effusions. An abdominal CT scan was normal.
|
|
The patient was admitted to the coronary care unit and
underwent diagnostic pleural aspiration and
pericardiocentesis. One liter of amber-colored pericardial fluid
was aspirated. Results of the analysis of pleural and
pericardial fluid appear in Table 3. A
diagnostic procedure was performed.
|
|
Hemodynamic Measurements During Pericardiocentesis (Ping Fai Wong, MD) |
|---|
|
TopCase Presentation (Ramesh...Hemodynamic Measurements During...Clinical Discussion (Eddy...Pathological Findings (Kent A....Clinical Follow-upFinal DiagnosisReferences |
|---|
A radial arterial catheter and a balloon-tipped pulmonary artery catheter were placed for hemodynamic monitoring before pericardiocentesis. An arterial blood pressure tracing showed that blood pressure decreased from 155/75 to 125/55 mm Hg during inspiration, which corresponded to a pulsus paradoxus of 30 mm Hg (Fig 1
).
The pulmonary artery catheter was inserted via the right
internal jugular vein. Pressure measurements were made in the right
atrium, right ventricle, main pulmonary artery, and the
pulmonary capillary wedge position while the catheter was being
advanced. Right atrial pressure was elevated, with an a wave of 17
mm Hg, a v wave of 14 mm Hg, and a mean pressure of 14 mm Hg. Early
diastolic y descent was absent. The right
ventricular pressure tracing showed a respiratory variation
ranging from 32/16 mm Hg during expiration to 40/18 mm Hg during
inspiration. The mean pulmonary capillary wedge pressure was 18
mm Hg. After the pericardial catheter was successfully placed, the
mean intrapericardial pressure measured 13 mm Hg. A
simultaneous mean pulmonary artery pressure
measured 15 mm Hg, nearly equivalent to the intrapericardial pressure.
One liter of serosanguinous fluid was removed from the pericardial
cavity. A subsequent blood pressure tracing showed resolution of the
pulsus paradoxus, and the mean intrapericardial pressure fell to -4
mm Hg (Fig 2).
|
|
|
Clinical Discussion (Eddy Barasch, MD) |
|---|
|
TopCase Presentation (Ramesh...Hemodynamic Measurements During...Clinical Discussion (Eddy...Pathological Findings (Kent A....Clinical Follow-upFinal DiagnosisReferences |
|---|
This was a middle-aged woman who presented to the hospital with cardiac tamponade and bilateral pleural effusions. She had low-grade fever, shortness of breath, and a pleuritic type of chest pain for several weeks before admission. The relevant blood laboratory data were hyponatremia, an abnormal albumin/globulin ratio, high haptoglobin, mild anemia with thrombocytosis, and a moderately elevated erythrocyte sedimentation rate. The serology report was notable for an antinuclear antibody titer of 1:80 with a speckled pattern, low total serum hemolytic complement, and negative anti-dsDNA antibody. The hepatitis B core antibody test was positive, and the hepatitis B surface antibody test was negative.
The patient's ECG showed sinus tachycardia and
low-voltage QRS complexes in the standard leads, with diffuse,
nonspecific ST-segment and T-wave changes. The chest roentgenogram
revealed bilateral pleural effusions and cardiomegaly. Unfortunately,
the first roentgenographic study was lost; Fig 3 shows
the chest roentgenogram taken after pericardiocentesis. The fluid
removed during pericardiocentesis had the characteristics of an exudate
with lymphocytosis. A pleural tap was performed and revealed a
transudate rich in lymphocytes.
|
A transthoracic echocardiogram showed cardiac chambers of
normal size and wall thickness. Global left and right
ventricular systolic function was increased. A
"swinging" heart movement resulted from a large amount of
pericardial fluid (Fig 4). Numerous fibrinous strands
were attached to the parietal pericardium. Early diastolic
right ventricular collapse (Fig 5) and late
diastolic right atrial free wall compression were observed.
The inferior vena cava was dilated, and its diameter did
not vary with respiration. On pulsed Doppler examination of the
mitral inflow, a 40% inspiratory decrease of peak E-wave velocity with
a corresponding 60% increase of peak E-wave velocity of the tricuspid
inflow was detected. Hepatic vein flow showed only x waves during
expiration and apnea. Together, these findings strongly suggest the
presence of cardiac tamponade.
|
|
Cardiac tamponade occurs when diastolic filling is impaired by an absolute rise in intrapericardial pressure. The diagnosis is based on clinical examination, which reveals elevated systemic venous pressure, tachycardia, dyspnea, and pulsus paradoxus. The presence of a paradoxical arterial pulsation is insufficient to diagnose cardiac tamponade, but its absence does not rule out the diagnosis. Pulsus paradoxus may be present in patients with chronic obstructive pulmonary disease, massive pulmonary embolism, constrictive pericarditis, or right ventricular infarction; these conditions may induce changes in both left and right ventricular filling that also depend on the respiratory phase. In the presence of cardiac tamponade, pulsus paradoxus may be absent if the patient (1) has severe left ventricular systolic dysfunction or severe aortic regurgitation in which left ventricular diastolic pressure may equal intrapericardial pressure, (2) has an atrial septal defect in which the respiratory changes in ventricular filling are significantly diminished, or (3) is being mechanically ventilated, in which the inspiratory increase in intrathoracic pressure may abolish the mechanism that gives rise to pulsus paradoxus.1 2
The echocardiographic signs of cardiac tamponade are
early diastolic collapse of the right
ventricular free wall, late diastolic
compression of the right atrium, and persistently dilated venae cavae.
On pulse-wave Doppler, one can observe a decline on inspiration
of 40% of peak early diastolic filling wave (E wave)
velocity of the mitral inflow, an 80% increase in peak E-wave velocity
of the tricuspid inflow, and an 85% increase in the isovolumetric
relaxation time of the left ventricle. Furthermore, y waves may be
absent from hepatic vein flow during apnea or expiration. Right
ventricular hypertrophy, changes in the right
ventricle and/or atrium compliance, or an increase in intravascular
volume may abolish these Doppler echocardiographic
signs.3 Low-pressure tamponade with hypovolemia and
localized tamponade, especially in the period immediately after cardiac
surgery, are distinct categories.
The fluid in the patient's pericardial and pleural spaces had probably been accumulating for several weeks before her admission to the hospital. Because of the amount of pericardial fluid, a recent acute event, such as an acute myocardial infarction with acute or subacute rupture of the left ventricular free wall, was unlikely. A slowly developing process is more likely to have enabled the accumulation of a large amount of fluid in the distensible pericardial sac while simultaneously producing the slow development of symptoms. Rapid accumulation of even 200 mL of fluid may produce cardiac tamponade. Other causes for tamponade, such as cardiac surgery, chest trauma, or the rupture of an ascending aortic aneurysm, were excluded by the patient's echocardiographic data and medical history.
A primary or secondary malignant process may induce the formation of a large pericardial effusion. However, primary pericardial malignant tumors are rare,4 and their solid or cystic structure was not identified by echocardiography in this patient. Lung, breast, or gastrointestinal carcinoma; melanoma; leukemia; and Hodgkin's or non-Hodgkin's lymphoma may involve the pericardium and are the most frequent causes of massive pericardial effusion.5 Lung carcinoma with pleural and pericardial involvement would already have been in an advanced stage, but the patient had no symptoms or signs of this disease. In addition, the chest roentgenogram did not disclose any pulmonary masses. The patient had no palpable breast lumps or axillary and/or supraclavicular adenopathy. The chest roentgenogram and negative CT scans of the chest and abdomen revealed no evidence of lymphomas. Because the patient was Chinese, hepatoma should be considered among the potential malignancies because of its increased prevalence, secondary to hepatitis B infection, among Asians. However, her symptoms and the results of her liver function tests made this diagnosis unlikely.
Pericarditis is the presenting symptom in 6% of women with
systemic lupus erythematosus
(SLE).6 Except for serositis, however, no symptoms in this
patient suggested SLE. In addition, the patient denied taking
lupus-inducing drugs such as hydralazine or
procainamide. Antinuclear antibodies are very sensitive but not
specific for SLE; they may also increase with age or with viral or
chronic infections. The patient's speckled antinuclear antibody
pattern reflected the presence of antibodies directed against non-DNA
nuclear constituents. The anti-dsDNA antibody test, which is
relatively specific for SLE, was negative. The anti–extractable
nuclear antigen assay detects antibodies against two extractable
nuclear antigens: the Sm and RNP antigens.7 Antibodies
against the Sm antigen are characteristic for SLE, and a high titer of
anti-RNP antibodies is the hallmark of mixed connective tissue disease,
in which pleuropericardial involvement occurs in 25% of patients
and pericarditis is the most common manifestation.8
Pericardial effusion is also found in 40% of patients with progressive
systemic sclerosis,9 but no clinical or laboratory data
support such a diagnosis in this patient. Other diseases in which
pleuropericarditis may be a presenting symptom, such as rheumatic
fever or rheumatoid arthritis, are also unlikely in her case.
Metabolic disorders such as renal failure and myxedema, two
potential causes of massive pericardial effusion, are essentially ruled
out by the normal kidney and thyroid function tests.
Could this woman have sarcoidosis? The disease is not prevalent among Asians, and the pericardium is rarely affected. Most patients are <40 years old, and the lungs are almost always involved. The patient also lacks the multisystem organ manifestations that are characteristic of this disease.
The massive pericardial effusion in this patient may have been caused by an infectious agent. Viral infections, especially coxsackievirus B and echovirus 8, may cause pericarditis with clinical features that cannot be distinguished from idiopathic pericarditis. The diagnosis of viral infection is strongly supported by the greater than fourfold rise in serial neutralizing viral antibody titers during the initial 3 weeks of the illness. Pericardial effusion and constrictive pericarditis are potential consequences of such an infection. The symptoms of this patient, however, would be unusual in a benign viral infection. Pericardial effusion is one of the most frequent cardiac manifestations in AIDS patients, but the patient was HIV negative.
Among fungal infections, histoplasmosis is most often associated with cardiac tamponade,10 but it usually resolves spontaneously within 2 weeks. The patient's record does not indicate that she lived or traveled in an area where histoplasmosis is endemic, and her other symptoms and laboratory data do not support such a diagnosis. Other fungal diseases, such as blastomycosis, coccidioidomycosis, aspergillosis, and candidiasis, are usually lethal in a relatively short time.
Tuberculosis is the cause of 7% of cardiac tamponade cases, and it should be considered in any case in which pericarditis is not rapidly self-limiting.11 In the United States, between 1985 and 1990 the tuberculosis case rate increased by 15.8%. A third of the world's population is infected,12 and women are more frequently affected than men. Patients who have tuberculous pericarditis usually have a focus of infection elsewhere in the body. In a recent retrospective study, the main symptoms of tuberculous pericarditis were chest pain, dyspnea, cough, weight loss, and increased perspiration.13 The reported clinical signs, in order of decreasing frequency, were fever, pericardial friction rub, pericardial effusion, cardiac tamponade, and pleural effusion.
Tuberculous pericarditis is associated with pleural effusion in 61% of patients.13 What suggested tuberculous pericardial effusion in our patient? The patient may have had a reactivated infection of tuberculosis contracted many years ago in China. She might also have contracted tuberculosis during her air travel or her stay in the Middle East. Transmission of Mycobacterium tuberculosis has been associated with air travel,14 and there has been a resurgence of tuberculosis in both the United States and the countries of the Middle East. The patient's low-grade fever and loss of appetite and the laboratory data that suggest a chronic inflammatory or infectious process, together with the predominance of lymphocytes in both pleural and pericardial fluid, might point toward tuberculosis as the cause of her cardiac tamponade.
Serum hyponatremia may result from low antidiuretic hormone secretion or adrenal involvement by the infectious process. A morning serum cortisol level should help to rule out adrenal insufficiency. The relatively high glucose concentration in both pericardial and pleural fluid is an unusual finding in tuberculosis. A purified protein derivative test was not done. The test is positive in 40% of patients with idiopathic pericarditis and negative in 25% of patients with tuberculous pericarditis.11 An adenosine deaminase level >45 U/L in both the pericardial and pleural fluids might suggest tuberculosis. False-positive results have been described in neoplastic pericarditis. In our patient, adenosine deaminase was not measured. Today, available polymerase chain reaction techniques have a 97% accuracy and enable a diagnosis of tuberculosis in 48 hours by amplification of DNA from as little as 1 µL of pericardial fluid15 16 ; however, such a test was not done in our patient.
Pericardioscopy and thoracoscopy followed by pericardial and epicardial biopsy are relatively new techniques for diagnosing the cause of pericardial effusion. Protuberances on the epicardium are specific for tuberculous pericardial effusion and neoplastic disorders. The pericardial fluid seldom yields tubercle bacilli during early phases of the disease: <50% of patients have positive acid-fast stains.17 18 In nearly 100% of the cases in which tubercle bacilli are found in the pericardial tissue, tuberculosis is the diagnosis.15 19 The presence of caseating granuloma is nonspecific, because it is also found in fungal infections.
Because the incidence of constrictive pericarditis in cardiac tamponade cases is between 30% and 57% within 2 to 4 months of diagnosis, if the tamponade recurs after 4 to 6 weeks of therapy, pericardiectomy is indicated.13 18 The use of corticosteroids is controversial and apparently does not decrease the incidence of constriction but was reported to decrease mortality by threefold, from 14% to 3%.20
In summary, I believe that this patient had cardiac tamponade, probably caused by tuberculosis; a malignant process is a possible but less likely cause.
|
Pathological Findings (Kent A. Heck, MD) |
|---|
|
TopCase Presentation (Ramesh...Hemodynamic Measurements During...Clinical Discussion (Eddy...Pathological Findings (Kent A....Clinical Follow-upFinal DiagnosisReferences |
|---|
The diagnostic procedure was a pericardial biopsy. The specimen consisted of small, irregular fragments of soft tissue that measured, in aggregate, 3.2 cm. The biopsy demonstrated marked thickening of the pericardium, with the innermost surface layered with extensive granulation tissue accompanied by lymphohistiocytic and plasmacytic infiltrates (Fig 6
). Well-formed
caseating granulomata occupied the central portion of the specimen (Fig 7). Special stains confirmed the presence of
acid-fast organisms in peripherally localized
histiocytes. The histopathologic features are indicative of
granulomatous pericarditis secondary to acid-fast bacilli,
consistent with Mycobacterium tuberculosis. Because
microbiological cultures for Mycobacterium failed to grow
organisms, further characterization of the subtype of
Mycobacterium could not be established.
|
|
Tuberculous pericarditis may be histologically
classified into three stages: acute, subacute, and chronic. The
histopathologic features in this patient correspond to the subacute
stage, in which granulomatous inflammation is a predominant feature.
Caseation necrosis is usually encountered and is surrounded by
Langhans'-type histiocytes. Occasionally, caseous necrosis may
be exuberant and cover the surface of the heart. Acid-fast stains
and cultures are more likely to be confirmatory during the subacute
stage. Constrictive pericarditis, which has an incidence of 30% in
this disorder,21 may arise in either of the two latter
stages.
In one report, pericardial biopsy yielded positive results in 84% of patients with positive pericardial fluid microbiological cultures.20 Other reports record a lower diagnostic yield.19 22 The differences between the results of these studies may be partially attributed to the stage at which biopsies were obtained. The efficacy of polymerase chain reaction for confirming tuberculous pericardial effusions is being investigated, and this test may serve as a further adjunct in diagnosis.
|
Clinical Follow-up |
|---|
|
TopCase Presentation (Ramesh...Hemodynamic Measurements During...Clinical Discussion (Eddy...Pathological Findings (Kent A....Clinical Follow-upFinal DiagnosisReferences |
|---|
The patient was treated with a combination of isoniazid, rifampicin, ethambutol, and pyrazinamide; she has nearly completed this antituberculous therapy. She improved significantly both subjectively and objectively and returned to her usual activities.
|
Final Diagnosis |
|---|
|
TopCase Presentation (Ramesh...Hemodynamic Measurements During...Clinical Discussion (Eddy...Pathological Findings (Kent A....Clinical Follow-upFinal DiagnosisReferences |
|---|
Tuberculous pericarditis.
|
References |
|---|
|
TopCase Presentation (Ramesh...Hemodynamic Measurements During...Clinical Discussion (Eddy...Pathological Findings (Kent A....Clinical Follow-upFinal DiagnosisReferences |
|---|
-
Shabetai R, Fowler NO, Guntheroth WG. The
hemodynamics of cardiac tamponade and constrictive
pericarditis. Am J Cardiol. 1970;26:480-489. [Medline]
[Order article via Infotrieve]
-
Reddy SP, Curtiss EL. Cardiac
tamponade. Cardiol Clin. 1990;8:627-637.[Medline]
[Order article via Infotrieve]
-
Fowler NO. Cardiac tamponade: a clinical or an
echocardiographic diagnosis?
Circulation. 1993;87:1738-1741.
[Abstract/Free Full Text]
-
Pascual MA, Povar J, Munoz JR, Cabeza FG, Portoles A,
Casado IL, Lore B, Ibbara F. Pericardial mesothelioma.
Rev Esp Cardiol. 1989;42:559-561. [Medline]
[Order article via Infotrieve]
-
Mukai K, Shinkai T, Tominaga K, Shimosato Y.
The incidence of secondary tumors of the heart and pericardium:
a ten-year study. Jpn J Clin Oncol. 1988;18:195-201.
[Abstract/Free Full Text]
-
Fowler NO. The Pericardium in
Health and Disease. Mt Kisco, NY: Futura; 1965.
-
Fye KH, Sack KE. Rheumatic diseases.
In: Stites DP, Stobo JD, Wells JV, eds. Basic and Clinical
Immunology. Norwalk, Conn: Appleton & Lange; 1987:356-384.
-
Prakash UBS, Luthra HS, Divertie MB.
Intrathoracic manifestations of mixed connective tissue
disease. Mayo Clin Proc. 1985;60:813-821. [Medline]
[Order article via Infotrieve]
-
Smith JW, Clements PJ, Levisman J, Furst D, Ross M.
Echocardiographic features of progressive
systemic sclerosis: correlation with hemodynamic and
postmortem studies. Am J Med. 1979;66:28-33. [Medline]
[Order article via Infotrieve]
-
Wheat LJ, Stein L, Corya BC, Wass JL, Norton JA, Grider
K, Slama TG, French ML, Kohler RB. Pericarditis as a
manifestation of histoplasmosis during two large urban
outbreaks. Medicine. 1983;62:110-119. [Medline]
[Order article via Infotrieve]
-
Permanyer-Miralda G, Sagristá-Salueda J,
Soler-Soler J. Primary acute pericarditis: a prospective series
of 231 consecutive patients. Am J Cardiol. 1985;56:623-629. [Medline]
[Order article via Infotrieve]
-
Daniel TM. Tuberculosis. In: Isselbacher
KJ, ed. Harrison's Principles of Internal Medicine. 13th
ed. New York, NY: McGraw-Hill; 1994:710-718.
-
Sagristá-Salueda J, Permanyer-Miralda G,
Soler-Soler J. Tuberculous pericarditis: ten years' experience
with a prospective protocol for diagnosis and treatment.
J Am Coll Cardiol. 1988;11:724-728. [Abstract]
-
Driver CR, Valway SE, Morgan M, Onorato IM, Castro KG.
Transmission of Mycobacterium tuberculosis associated with air
travel. JAMA. 1994;272:1031-1035. [Abstract]
-
Maisch B. Pericardial diseases, with a focus on
etiology, pathogenesis, pathophysiology, new diagnostic
imaging methods, and treatment. Curr Opin Cardiol. 1994;9:379-388. [Medline]
[Order article via Infotrieve]
-
Seino Y, Ikeda U, Kawaguichi K, Yamamoto K, Sekiguchi
H, Nakayama T, Tkewaki S, Okubo A, Nagai R, Yyazaki Y, Schimada K.
Tuberculous pericarditis presumably diagnosed by polymerase
chain analysis. Am Heart J. 1993;126:249-251. [Medline]
[Order article via Infotrieve]
-
Fowler NO, Manitsas GT. Infectious
pericarditis. Prog Cardiovasc Dis. 1973;16:323-328. [Medline]
[Order article via Infotrieve]
-
Fowler NO. Tuberculous pericarditis.
JAMA. 1991;226:99-103.
-
Rooney JJ, Crocco JA, Lyons HA. Tuberculous
pericarditis. Ann Intern Med. 1970;72:73-78.
-
Strang JI, Kakaza HH, Gibson DG, Allen BW, Mitchison
DA, Evans DJ, Girling DJ, Nunn AJ, Fox W. Controlled clinical
trial of complete open surgical drainage and of prednisolone in
treatment of tuberculous pericardial effusion in Transkei.
Lancet. 1988;2:759-764. [Medline]
[Order article via Infotrieve]
-
Gleckman RA. Nonviral infectious
pericarditis. In: Spodick DH, ed. Pericardial
Diseases. Philadelphia, Pa: FA Davis; 1976:159-176.
-
Hageman GH, D'Esopo ND, Glenn WL. Tuberculosis
of the pericardium: a long-term analysis of forty-four
proved cases. N Engl J Med. 1964;270:327-332.
This article has been cited by other articles:
 |
|
 |
|
  G. Cherian, B. Uthaman, A. Salama, A. G. Habashy, N. A. Khan, and J. M. Cherian Tuberculous Pericardial Effusion: Features, Tamponade, and Computed Tomography Angiology, July 1, 2004; 55(4): 431 - 440. [Abstract] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||