History of Present Illness:

Ms C. is a 67 year-old female with past medical history significant for chronic bronchitis with frequent flares of acute bronchitis, secondary to extensive tobacco abuse, hypercholesterolemia and hypothyroidism. In 1994, both treadmill testing and echocardiography were normal.

She presented to her primary physician in early September 1999 with dyspnea, dyspnea on exertion and occasional nocturnal dyspnea. The presumed diagnosis at that time was acute bronchitis and she was treated as an outpatient with antibiotics. She returned approximately 7 days later with persistent and worsening dyspnea, and new complaints of chest pressure, "something pressing on my heart", stabbing back pain, abdominal distension and lower extremity edema. An electrocardiogram (ECG) revealed sinus rhythm with low voltage, a significant change from previous ECG, and nonspecific ST-T wave changes. She was referred to the pulmonology clinic at NCBH. Prior to her visit, chest radiography was performed which revealed bilateral pleural effusion, left greater than right, compressive atelectasis and cardiomegaly. With this information she was directly admitted to the Cardiology A service.

Social: Significant for > 100 pack-year history of tobacco use

Family Hx: Significant for non-premature CAD and hypertension.

VS: P: 72, R: 24, SBP: 128 with an additional 40 mm Hg pulsus paradoxus, DBP: 70

Neck: Supple without LA, TM, JVD, or bruit. The carotid upstrokes were brisk bilaterally

Chest: Decreased breath sounds at the bases with bilateral dullness to percussion left greater than right, mid lung ronchi and anterior wheezes

Cor: Regular rhythm with no palpable PMI or lift, heart tones were distant with S1 and S2 without definite murmurs, rubs, or gallups

Abd: Soft with right upper quadrant tenderness and 4 centimeters of palpable liver below the costal margin

Ext: Pulses were 2+ in the upper and lower extremities bilaterally. Palmar cyanosis was noted along with 2+ pitting edema below the knee

Sinus rhythm with a rate of 74, low voltage in both the limb and the precordial leads and nonspecific ST-T wave changes.

2d echocardiography revealed normal left ventricular chamber size and adequate LV performance. A moderate to large circumferential pericardial effusion was present with evidence of bi-atrial diastolic collapse without right ventricular diastolic collapse. Continuous wave doppler of the tricuspid and the mitral valve flow revealed no significant inspiratory/expiratory variation.

Pericardial effusion with cardiac tamponade.

Hospital Course:

She underwent pericardiocentesis, bronchoscopy with biopsy and pericardectomy and was unfortunately diagnosed with well differentiated squamous cell carcinoma of the lung.

Cardiac tamponade is a condition in which external compression on the walls of the heart results in severely reduced cardiac output. Hypotension and shock ensue despite maximal physiologic compensation with increased in heart rate and systemic vascular resistance. Tamponade most commonly occurs in the presence of hemodynamicaly significant pericardial effusion, however, large compressive bilateral pleural effusions (1) and pneumopericardium (2) have also been reported to produce tamponade physiology.

It is difficult to discern who first described cardiac tamponade. Kussmal first described Pulsus Paradoxus: the diminution or complete absence of the radial pulse during inspiration (3). It was felt to be a paradox because despite the absence of radial pulses, heart sounds could be auscultated. Experiments were initially conducted in the late 1800's (4) and early 1900's to induce cardiac tamponade in an attempt to understand the pathophysiology of the disease process. Initially mineral oil was injected into the pericardial space to generate tamponade and later saline became the infusate of choice. Several early theories were proposed as to the mechanism by which pleural effusion induced pulsus paradoxus in tamponade. However, Katz provided the first appropriate physiologic explanation of pulsus paradoxus and tamponade physiology (5). In short, it was proposed that a combination of decreased left ventricular filling and impingement of the interventricular septum on the left ventricular cavity were required for the development of pulsus paradoxus in cardiac tamponade. It was not until 1965 and the demonstration by Shabetai that when right ventricular cardiac output was kept constant that no amount of pericardial effusion or degree of cardiac tamponade would result in pulsus paradoxus.

The physiology of cardiac tamponade is illustrated in the highly stylized cartoon in Figure 1. adapted from Braunwald (6). In this figure the intrathoracic space is enclosed by the largest box while the intrapericardial space is shaded and contains the schematic heart. The figure can be subdivided into the phases of respiration, end expiration, inspiration and expiration. In the end expiratory phase there is equalization of diastolic pressures which is the hemodynamic hallmark of cardiac tamponade. With inspiration the intrathoracic pressure is markedly reduced and a "down hill " gradient between the central venous pressure and the pulmonary arteries develops. Blood readily flows along this gradient filling the right ventricle and causing impingement of the interventricular septum on the cavity of the left ventricle. Simultaneously, an "up hill" gradient between the pulmonary veins and the left atrium develops resulting in decreased venous return to the left heart. With decreased left ventricular volume; stroke volume, cardiac output and systolic pressure generated during inspiration is markedly reduced. At the end of inspiration the compliance mechanisms of the thoracic cage and the pulmonary parenchyma are stretched to a maximum.

Figure 1. Pathophysioloy and Hemodynamics of Cardiac Tamponade ______________________________________________________________________________________

At this point, with the initiation of expiration, the intrathoracic and pulmonary vascular bed pressure is greatly increased. Blood is essentially pushed into the left atrium and the left ventricle resulting in an increase in end diastolic volume and improved contractility and cardiac output. Concurrently, the increased intrathoracic pressure hinders blood return to the pulmonary arteries and the right heart. 

____________________________

Symptoms of Cardiac Tamponade:

Chest pain
Oppressive Precordial
Dyspnea
Apprehension
Cough
Dysphagia
Hoarseness
Singultus
Early Satiety
Nausea
Abdominal Pain
____________________________ 
Table 1. Symptoms of cardiac tamponade

Clinical Signs/Physical Exam Findings in Cardiac Tamponade

Table 2. lists common and not so common physical findings seen in patients with cardiac tamponade. In general patients appear to be ill and in shock. In Table 3. the data compiled by Guberman is presented (8). This is a compilation of the physical findings of 56 patients with cardiac tamponade diagnosed at the bedside. The most common of the physical findings present in his population was increased juglar venous pressure, which was observed in 100% of patients. Interestingly 98% of patients had a pulsus paradoxus greater than 10mm Hg while only 36% had hypotension (systolic blood pressure less than 100 mm Hg). Since this time several reports of low pressure tamponade have been published (9,10). In this condition cardiac tamponade is present and aggravated by intravascular volume depletion and JVD is absent. Therefore, patients with cardiac tamponade and without JVD likely are dehydrated and would benefit from the administration of intravenous fluids.

Physical Findings of Cardiac Tamponade

As previously mentioned Kussmaul first reported paradoxical pulse, the weakening or disappearance of the radial pulse with respiration. Early studies in which induced cardiac tamponade in anaesthetized dogs or in patients with tamponade and hypotension led to the adoption of a fixed number of 10 mm Hg as the criteria for hemodynamicaly significant pulsus paradoxus. However, more recently as described by Guberman, a majority of patients with cardiac tamponade were normotensive with SBP greater than 100mm Hg. Several authors argue, a more physiologic value for hemodynamically significant pulsus paradoxus would be greater 10% of the systolic pressure at which all the Korotkoff sounds have an even intensity (11). Table 4. presents a step by step method for obtaining and calculating the pulsus paradoxus.

______________________________________________________________________________________

Determination of Pulsus Paradoxus

______________________________________________________________________________________

1. Place the patient in a position of comfort and conduct manometric studies during baseline respiration.
2. Raise sphygmomanometer pressure until Korotkoff sounds disappear.
3. Lower pressure slowly until first Korotkoff sounds are heard during early expiration with their disappearance during inspiration
4. Record this pressure.
5. Very slowly lower pressure until Korotkoff sounds are heard throughout the respiratory cycle with even intensity.
6. Record this pressure.
7. The difference between the two recorded pressures is the Pulsus Paradox.
8. Hemodynamically significant pulsus paradox is greater than or equal to 10% of the pressure at which all Korotkoff sounds are heard with even intensity.

Although pulsus paradoxus has been and continues to be heavily relied upon to diagnose cardiac tamponade, in several pathologic conditions pulsus paradoxus may not be present depite the presence of cardiac tamponade. For example: atrial and ventricular septal defects with intracardiac shunting; conditions with fixed cardiac output such as supravalvular, valvular and sub valvular aortic stenosis, mitral valve stenosis, and in conditions with severely impaired left ventricular systolic function such as severe cardiomyopathy or acute myocardial infarction.

______________________________  

Etiologies of Pulsus Paradoxus 

Large pulmonary embolus
Severe COPD exacerbation
Labored respiration
Constrictive pericarditis
Restrictive cardiomyopathy
Right ventricular infarction
Circulatory shock
Large pleural effusions
Tense ascites
Extreme obesity

______________________________ 
Table 5. Etiologies of pulsus paradoxus other Pericardial effusion and cardiac tamponade.

______________________________ 

Etiologies of Electrical Alternans 

Soon after World War II and the advent of radar technology, doppler technology was adapted for use in humans. In 1954, Edler made the first reference to the use of the ultrasonic reflectoscope in the diagnosis of pericardial effusion (13). Soon thereafter technologic advances were made and M mode echocardiography evolved. Echocardiography has become the gold standard in the diagnosis of pericardial effusion. In 1974, Horowitz evaluated the M mode echocardiography in the diagnosis and evaluation of Cardiac tamponade (14). Although very useful, M mode technology proved somewhat limited in identifying the hemodynamic significance of some pericardial effusions. In the mid 1970s 2 dimensional (2D) echocardiography was being developed. Schiller, in 1977, reported the use of 2D echocardiography in the diagnosis of cardiac tamponade (15). 2D echo technology has improved substantially since then and is a mainstay in the evaluation of pericardial effusions with the assistance of M mode echo. Further advancements in doppler technology to assess blood flow velocities allowed Pandian, in 1985, to identify significant respiratory variation in the flow velocities across the mitral and tricuspid valves in patients with cardiac tamponade (16).

I am limiting the discussion of echocardiography and echocardiographic findings to those seen in 2D and pulsed wave doppler echocardiography. Figure 2. presents the most commonly seen 2D echocardiographic views and the views that are most helpful in the assessment of pericardial effusions.

Figure 2. Common 2 Dimensional Echocardiographic views ______________________________________________________________________________________

______________________________________________________________________________________

Using 2D echo with M mode assistance, the following findings have been found to be useful in the assessment of patients with hemodynamicaly significant pericardial effusions and patients with cardiac tamponade: right atrial collapse, right ventricular diastolic collapse and inferior vena cava (IVC) plethora (the percentage of decrease in the diameter of the IVC with deep inspiration). Pulse wave doppler technology allows the interrogation of the blood flow velocities across all valves, but most importantly, across the tricuspid and the mitral valves in the assessment of the hemodynamically significance of pericardial effusions. Referring to Figure 1, one can readily see that during inspiration the flow velocities across the tricuspid valve increase while concomitantly decreasing across the mitral valve and vice-versa with expiration as blood is being forced into the left atrium and left ventricle.

Unfortunately, although there is data available assessing the sensitivity and specificity of the echocardiographic findings of right atrial collapse, right ventricular diastolic collapse and respiratory flow variation across the mitral and tricuspid valves to diagnose cardiac tamponade, it is largely retrospective, poorly blinded, if blinded at all, and small numbers of patients are included in each study. Further, echocardiography is an excellent mode of imaging and obtaining data about cardiac function and pathologic states however the interpretation of the data can be highly subjective and is greatly influenced by experience. As with many studies, these were data collected in tertiary referral centers. The echocardiograms were performed by highly experienced and specialized technologists and then were interpreted by highly experienced and motivated cardiologists. These factors in and of themselves may make the results difficult to generalize to non-tertiary referral centers. Further, even prospective studies can have significant bias especially by the technologists who may make that extra effort to examine certain parameters which may otherwise be glossed over in non study patients. Therefore, I will not provide indepth analysis of the data and individual manuscripts, but rather, I will present the evolution of the techology and echo findings, some data as to the sensitivities and specificities of various echo findings for the diagnosis of cardiac tamponade.

In 1977, Schiller identified a novel finding in patients with pericardial effusion and cardiac tamponade. This finding was right ventricular diastolic collapse. In an attempt to more clearly delineate the potential role of this finding in the identification of patients with cardiac tamponade, a chart review of patients with cardiac tamponade as diagnosed by clinical criteria was performed. The echocardiograms of 17 patients were read and 16 of the 17 patients with clinically diagnosed tamponade were identified as having right ventricular diastolic collapse. The one patient with tamponade who did not have right ventricular diastolic collapse was noted to have severe COPD. These findings suggest that right ventricular diastolic collapse in the presence of pericardial effusion may identify patients with hemodynamically significant effusions in the absence of severe COPD (15).

Soon thereafter in 1983, Gilliam, using improved 2D echocardiographic imaging identified right atrial collapse in patients with cardiac tamponade (17). In an attempt to determine the sensitivity and specificity of the finding of right atrial collapse in identifying patients with cardiac tamponade the echocardiograms of 104 patients with moderate and large pericardial effusions were compared with the echocardiograms of 19 patients with clinically diagnosed cardiac tamponade. When the echocardiograms were assessed for any degree of right atrial collapse the finding was found to have a sensitivity of 100% and a specificity of 84%. However, if the duration of right atrial collapse was increased to greater than 34% of the cardiac cycle length, the sensitivity remained 100% and the specificity increased to 100%. These results suggest that prolonged right atrial collapse was an excellent marker of cardiac tamponade.

Singh followed in 1984, looking at a head to head comparison of the sensitivity and specificity of right ventricular diastolic collapse and right atrial collapse in patients with pericardial effusion referred for pericardiocentesis (18). All patients underwent invasive hemodynamic monitoring of right atrial, intra pericardial and pulmonary cappilary wedge pressure and 13 patients were found to have cardiac tamponade Three patients had no evidence of tamponade physiology. Of the 13 with tamponade, patients one was found to have pulmonary hypertension and this patients echocardiogram revealed neither right atrial or right ventricular diastolic collapse. Of the remaining 12 patients with tamponade, all 12 had right ventricular diastolic collapse while only 9 had right atrial collapse. Of the 3 patients without cardiac tamponade none had right atrial or right ventricular diastolic collapse.

Taking an opposite approach, Levine prospectively identified 50 patients with pericardial effusion and either right atrial or right ventricular diastolic collapse. Thorough physical examinations were performed prior to pericardiocentesis. Postpericardiocentesis all patients had some degree of hemodynamic improovement. The physical findings in the patients with pericardial effusion and right heart collapse were compared with the physical findings in patients diagnosed with cardiac tamponade at the bedside as published by Guberman (8). Overall, the incidence of positive physical findings in the patients with right heart collapse was far less than in the population of patients with bedside diagnosis of tamponade. Overall the patients seemed to be far less ill. With these results and hemodynamic improvement post pericadiocentesis the authors proposed that right atrial or right ventricular diastolic collapse might be evidence enough to proceed with urgent pericardiocentesis preventing the development of frank cardiac tamponade.

More recently, pulse wave doppler technology came of age and Appleton compared the flow velocities across the mitral and the tricupid valves in 21 patients with pericardial effusion and 21 matched controls without pericardial effusion (20). Of the 21 patients with pericardial effusion, 7 had cardiac tamponade by physical exam and invasive hemodynamic monitoring. The results are shown in Table 7.

In this series, the variation of flow velocities across the mitral and tricuspid valves appears to be graded. As would be expected, there is slight variation of flow velocity with respiration in normal controls. Interestingly, a small population of patients with pericardial effusion had slight variation in flow velocity with respiration, while others had increased variation in flow velocities but less than that seen in patients with cardiac tamponade. It is known there is a continuum of patients with pericardial effusion from hemodynamically compensated to decompensated cardiac tamponade (21, 22). These results indicate that the degree of respiratory variation of flow across the mitral and tricuspid valves may help noninvasively grade the degree of hemodynamic compromise and direct therapy.

The most common etiology of pericardial effusion resulting in cardiac tamponade is malignancy. Approximately one-third are the result of malignancy. With increasing numbers of catheter based interventions and sternotomy with pericardotomy the incidence of tamponade in these cases is on the rise. Other less common etiologies of pericardial effusion resulting cardiac tamponade include: idiopathic, uremia, connective tissue diseases, bacterial infections, post-infarction rupture and myxedema.

• The gold standard for the diagnosis of pericardial effusion is echocardiography.
• The diagnosis of CARDIAC TAMPONADE is based solely on PHYSICAL EXAM.
• There is a continuum of patients with pericardial effusion, those with a compensated hemodynamic status and those with a decompensated tamponade status. The point at which the compensatory mechanisms fail and tamponade ensues may not be readily identified by echo, but usually can be readily identified by physical exam.
• Several pathologic conditions such as right ventricular hypertrophy or increased right heart pressure, valvular abnormalities, septal defects etc. may result in no echocardiographic findings of hemodynamic compromise, despite its presence in patients with pericardial effusion.