CO is the product of stroke volume (SV) and HR: CO = HR × SV. Therefore, a low CO state can occur when SV and/or HR decrease. Fast HRs, of either ventricular or atrial origin (i.e., atrial fibrillation), may also cause a low CO state because the diastolic filling time is decreased or the atrial contribution to the ventricular filling is reduced or abolished. Because SV is the difference between left ventricular end-diastolic volume (LVEDV) and left ventricular end-systolic volume (LVESV), the etiology of a reduced SV can be more accurately determined by exploring whether there has been a change in the preload (LVEDV), the afterload, and/or the left ventricular contractile state. Therefore, increased (that may cause left ventricular overdistention and systolic dysfunction) or decreased preload can also lower the CO. Myocardial contractility can be compromised by either direct (myocardial ischemia, infarction, stunning, or hibernation) or indirect causes (mechanical factors as in cardiac tamponade, pulmonary embolism; pharmacologic agents; and toxins, such as inflammatory cytokines and bacterial endotoxins). A combination of different hemodynamic disturbances can coexist in the same patient. For example, a patient can have both a decrease in contractility and a reduced preload to account for the low CO state. Therefore, correcting one problem alone may not produce the optimal hemodynamic condition.

Cardiac tamponade is defined as significant extrinsic compression of the heart by accumulating intrapericardial blood and clots (postcardiotomy, dissecting aortic aneurysm, trauma, anticoagulant therapy), exudative effusions (malignant states, infective pericarditis, idiopathic pericarditis), nonexudative effusions (uremia, systemic lupus erythematosus, rheumatoid arthritis, idiopathic, radiation), and air. If the intrapericardial pressure is increased enough, globally or regionally, cardiac tamponade may occur irrespective of the actual intrapericardial volume. Cardiac tamponade is a pathophysiologic continuum that on one extreme may be clinically insignificant and on the other extreme present as a life-threatening condition requiring emergent surgical attention. In clinical tamponade, as the pericardial pressure approaches the intra-atrial and intraventricular pressures, the SV progressively declines, thus resulting in systemic hypotension and cardiogenic shock.

In the setting of cardiac surgery, acute cardiac tamponade can occur over minutes, hours, or after a few days postoperatively, with a reported incidence of 0.5% to 5.8%. In the typical patient, a significant chest tube output (greater than 200 mL per hour) in the immediate postoperative period, with or without hemodynamic signs of inadequate cardiac output, is a sign of an increased amount of blood around the heart. Alternatively, the chest tubes may become clogged by blood clots impeding mediastinal drainage and the clinical picture of cardiac tamponade develops sooner.

Delayed tamponade has been arbitrarily defined as cardiac tamponade occurring 5 to 7 days after pericardiotomy and its incidence is 0.3% to 2.6%. Delayed tamponade is often misdiagnosed because of a low index of suspicion and clinical signs and symptoms that are confused with congestive heart failure, pulmonary embolism, and generalized fatigue (i.e., failure to thrive postoperatively). The most common culprit in delayed cardiac tamponade is anticoagulant therapy with warfarin (Coumadin), heparin, or platelet-inhibiting agents. However, a pericardial effusion after cardiac surgery is common. Using echocardiography, Weitzman et al. evaluated 122 consecutive patients postoperatively and demonstrated that most of them (103 of 122, 84%) had pericardial effusions. Generally, effusions reach their maximum size by the 10th postoperative day and regress spontaneously thereafter.

The diagnosis of cardiac tamponade following cardiac surgery is often difficult to make and requires a high degree of clinical suspicion, physical examination, proficient knowledge of PAC-derived measurements, and diagnostic tools such as echocardiography and chest radiography. Often, the diagnosis is made in the operating room, whereby evacuation of clot and blood from the mediastinum is associated with a drastic improvement in hemodynamics (Fig. 12.1). Relying on one modality alone for diagnosis can lead to inaccurate management decisions and increase patient morbidity. For example, the PAC data can be misleading postcardiotomy because the classical teaching of equalization of diastolic blood pressures in cardiac tamponade (CVP = pulmonary artery diastolic pressure [PAD] = PAOP) is infrequently observed because the heart is not surrounded by the transected pericardium. Thus, blood or clot does not distribute homogeneously around the heart for the pericardial diastolic pressures to equalize.
Generally, the CVP is elevated with significant compression of the right heart, but reduced blood flow to the left heart may result in low left-sided heart pressures that would be inconsistent with “classical” tamponade physiology. The majority of patients with postcardiotomy cardiac tamponade have atypical clinical, hemodynamic, and/or echocardiographic findings, mostly because the compression is localized. Therefore, the diagnosis of postoperative cardiac tamponade should be considered every time hemodynamic deterioration is encountered, particularly when reduction in cardiac output or blood pressure or both are not readily responsive
to conventional management. Furthermore, in delayed tamponade, the complaints tend to be vague and invasive hemodynamic data (i.e., PAC) is usually not available to allow prompt diagnosis of tamponade. Whenever the patient is not progressing as expected postoperatively and signs of end-organ dysfunction (fall in urine output or increase blood urea nitrogen and creatinine) are present, an echocardiogram should be obtained to rule out cardiac tamponade.

Regional cardiac tamponade occurs when one or more cardiac chambers (and not necessary the entire heart) become compressed by blood or blood clot or both, thereby compromising heart function (Figs. 12.2 and 12.3). Postoperatively, a right atrial hematoma often becomes localized around the anterior and lateral walls. Clots can also be found behind the left atrium at the level of the oblique sinus. Postcardiotomy, the regional collapse of the right atrium or the right ventricle in diastole, is the most common echocardiographic finding in “early” cardiac tamponade. Selective compression of the right heart by hematoma becomes less prominent in “delayed” tamponade, as the right heart becomes adherent or tethered to
the anterior chest wall. The diagnosis of regional tamponade is often misdiagnosed because the classical features of tamponade are often absent and blood and blood clots are unevenly distributed around the heart. The clinical presentation may be mistaken for congestive heart failure, acute left or right ventricular dysfunction, septic shock, or pulmonary embolism.

Normally, during spontaneous inspiration, the extrathoracic to intrathoracic pressure gradient is increased and the filling of the right heart is slightly larger than the filling of the left heart. In the latter, the decreased intrapulmonary pressure during inspiration will cause a relative “pooling” of the blood in the lungs and decreases its filling gradient. That is associated with an inspiratory decrease of less than 10 mm Hg in the arterial systolic pressure along with an accompanying inspiratory decrease in the CVP. A paradoxical pulse differs from the normal situation in two aspects: the inspiratory fall of the arterial pressure exceeds 10 mm Hg and the inspiratory venous pressure remains steady or increases (Kussmaul sign), instead of decreasing. Echocardiographic studies of patients with cardiac tamponade by D’Cruz et al. described phasic respiratory changes whereby both left ventricular dimensions and mitral valve excursion decreased during inspiration. In contrast, right ventricular dimensions increased in association with a shifting of the interventricular septum toward the left ventricle. That is, the right heart filling is done at the expense of the left. Of importance, pulsus paradoxes and the phasic respiratory changes in ventricular dimensions and systemic systolic arterial pressure can also be present in respiratory distress, airway obstruction, chronic obstructive pulmonary disease, and pulmonary embolism, but they may be absent when patients cannot generate sufficient negative intrapleural pressure during inspiration, as in chest wall trauma, neuromuscular disease, and pneumothorax. Similarly, patients under positive pressure mechanical ventilation or with severe aortic regurgitation do not exhibit pulsus paradoxus.

Under normal conditions, the average SV of the right ventricle equals the SV of the left ventricle; however, cyclical respiratory differences in left and right ventricular SV do occur. During inspiration, the generated negative intrapleural pressure facilitates blood return to the right heart, whereas blood return to the left heart is diminished for the following reasons: (1) lung expansion increases its reservoir for blood and (2) the increase in right ventricular filling causes the interventricular septum to “bulge” leftward, thereby reducing left ventricular dimensions and altering its compliance and filling (ventricular interaction). During exhalation, the reverse process occurs. In the case of cardiac tamponade, the ventricular interaction is augmented and involves not only the interventricular septum but also other chambers depending on the etiology of the tamponade (fluid vs. clot, regional vs. global tamponade). Fundamentally, as intrapericardial content increases, it will reach a point whereby the parietal pericardium cannot stretch to accommodate the rising pressure. Because of the fixed space within the pericardium, the cardiac chamber dimensions become smaller as the pericardial contents increase. First, the thin-walled and more compliant atria progressively get compressed, followed by the right ventricle, and finally by the left ventricle.
Overall, the physiology of ventricular interaction in clinical cardiac tamponade becomes complex as the pressure-volume relation of mediastinal and chest structure is altered with each heartbeat and from respiratory and neuroendocrine influences.

In the absence of coronary artery disease, coronary blood flow is reduced in cardiac tamponade, but this reduction is not sufficient to add an ischemic insult to the heart because there is a proportionate decrease in ventricular preload (underfilled heart) and ventricular afterload. Consequently, both myocardial work and oxygen consumption are reduced. In contrast, patients with coronary artery disease may be at increased risk for myocardial ischemia and infarction.

Electrocardiographic changes in the setting of cardiac tamponade include nonspecific ST-T wave abnormalities, low-voltage QRS complex, signs of myocardial ischemia and pericarditis, and electrical alternans (Fig. 12.4). The electrocardiographic change of electrical alternans may reflect a hemodynamic pathology rather than an anatomic abnormality and is not very specific (very few patients with tamponade present with electrical alternans).

On standard anterior-posterior chest radiography, the cardiac silhouette may appear normal in size or extremely enlarged, depending on the acuity and chronicity of the tamponade process. Normally, the pericardial fluid amounts to 15 to 25 mL. Acute tamponade physiology can arise with as little as 150 mL of effusion, whereas chronic effusions may exceed 1,000 mL before clinical signs and symptoms of cardiac tamponade become evident. In the latter case, on chest radiography film, the cardiac silhouette will appear “widened” with or without features such as obscuring of the pulmonary vessels at the hilum and a globular or “water bottle” configuration of the heart.

Transthoracic or transesophageal echocardiography can differentiate between cardiac dysfunction arising from direct myocardial injury and cardiac dysfunction associated with mechanical processes. Pericardial effusions can be readily seen and semi-quantitated. Likewise, pericardial blood clots can be demonstrated to compress a cardiac chamber. Additional echocardiographic signs observed in tamponade include (1) diminished left ventricular dimension and mitral valve excursion during spontaneous inspiration, (2) shifting of the interventricular septum toward the left ventricle, (3) fluctuation of transvalvular (mitral and aortic) flow seen by Doppler techniques, (4) diastolic expansion of the right ventricular chamber, and (5) in some cases, a systolic notch on the right ventricular epicardium (Figs. 12.2 and 12.3). Although many of the findings of both radiographic film and echocardiogram can be suggestive of cardiac tamponade, not a single sign is 100% sensitive and specific.

Significant bleeding following cardiac surgery requiring surgical exploration (“bring-back”) occurs in approximately 3% of cases. Often, the source of bleeding cannot be ascertained and its etiology is ascribed to a coagulation defect. Procedures that require CPB are known to adversely affect the hemostatic mechanisms, including vascular endothelium, fibrinolytic and coagulation factors, platelet, plasminogen, and inflammatory pathways (Fig. 12.5). Under normal
settings, these systems act in concert and in opposition with one another to prevent bleeding by promoting clot formation at the site of endothelial injury (procoagulant effects) and prevent or destroy clot formation (anticoagulant effects—either directly or indirectly through inhibitory mediators). The procoagulant system becomes activated once blood elements come in contact with a nonendothelialized surface (bypass circuit), resulting in consumption of platelets, coagulation factors, and fibrinogen (tissue factor pathway). The potential for postoperative bleeding will depend on the degree of activation and preexisting levels of the procoagulants as well as other factors such as the duration of CPB, systemic hypothermia, comorbid states (renal and liver insufficiency), type of surgery (reoperation, circulatory arrest, combined procedures), and prior or current treatment with anticoagulant (platelet inhibitors, warfarin process). Of greater importance is the loss of several membrane glycoproteins (IIb-IIIa) that interact with fibrinogen and other cellular elements, allowing for platelet aggregation and propagation of clot formation. Overall, postcardiotomy bleeding is due to (1) a qualitative and a quantitative platelet defect, (2) deficiency, and (3) surgical bleeding. The latter can arise from either venous or arterial sources that were undetected before closure of the chest. Attentive and meticulous effort to surgical hemostasis can significantly reduce the incidence of bring backs.