In normal subjects, the pericardium is difficult to visualize since the pericardial cavity is only a potential space and visceral and parietal pericardial layers appear as a single echo. In the setting of pericardial effusion, the fluid appears as a sonolucent area (or clear space) separating epicardium from pericardium.324 Pericarditis may be unaccompanied by pericardial effusion and in such cases may be undetectable by echocardiography. In addition, although thickening and/or calcification of the pericardium may be detectable by echocardiography in patients with constrictive pericarditis, cardiac ultrasound is limited in this capability. Therefore, the evaluation of constrictive pericarditis by echocardiography primarily involves Doppler flow recordings.325
Echocardiography is the diagnostic procedure of choice for detection of pericardial fluid (Fig. 15–135), and early M-mode studies demonstrated that volumes as small as 20 to 30 mL could be detected reliably.325 As both myocardium and pericardium are echo-reflective and pericardial fluid is not, a sonolucent area between the epicardium and pericardium is diagnostic of a pericardial effusion. Although epicardial-pericardial separation may be seen during systole in normal cases, separation throughout the cardiac cycle is abnormal.326 Descending aorta, coronary sinus, pleural effusion, pericardial cyst, and LV pseudoaneurysm occasionally may be mistaken for pericardial effusion.
Echocardiography can be used to identify pericardial loculations, fibrous strands, and pericardial tumors as well as to assess the size of effusions324 (Fig. 15–136). Pericardial effusions may be concentric or loculated (the latter type is especially common with postoperative, infective, and malignant effusions). As pericardial tissue reflects upon itself behind the LA between the pulmonary veins (the oblique sinus), fluid is rarely seen in this area. Small, nonloculated effusions may move depending on patient position and thus are often drawn posteriorly and inferiorly by gravity during routine imaging. A rim of pericardial fluid surrounding the heart is evidence of a moderate or large effusion, and the heart can sometimes be seen "swinging" back and forth within the pericardial space, creating the mechanism of electrical alternans. In general, small effusions are seen posteriorly rather than anteriorly on supine imaging. Moderate-sized (100 to 500 mL) nonloculated effusions are present both anterior and posterior to the heart. Large nonloculated effusions (>500 mL) are circumferential and frequently allow free motion of the heart within the fluid-filled space.
Distinguishing between pericardial and pleural effusions is occasionally difficult with echocardiography.327 If these conditions coexist, the pericardium usually can be identified as a linear density separating fluid in the two spaces. The parasternal long-axis view is often helpful in differentiating the disorders. The descending aorta is a mediastinal structure; therefore pericardial effusions will often separate the heart and descending aorta, while pleural effusions are seen inferior and posterior to the aorta327 (Fig. 15–137). In cases of large pleural effusions, atelectatic lung tissue also may be present (Fig. 15–137). Subcostal views are often valuable and may yield the only satisfactory transthoracic images in postoperative or posttraumatic cases. The inferior vena cava also can be imaged in this view; if the vessel does not display inspiratory collapse greater than 50 percent of its maximum diameter, elevated RA pressure is present.
On parasternal images, an echolucent space is sometimes visualized anterior to the RV.328 Although this finding may represent pericardial fluid, it usually is caused by epicardial fat (without effusion) and has no pathologic significance. Therefore the diagnosis of pericardial effusion based solely on the presence of this anterior clear space should be avoided.
As the pericardium is a relatively noncompliant membrane that adapts slowly to volume changes, pericardial effusions (especially those that accumulate rapidly) may limit cardiac filling and cause cardiac tamponade. Echocardiography can help diagnose this condition by detecting (1) morphologic signs of increased intrapericardial pressure and (2) abnormal intracardiac flow patterns caused by tamponade and enhanced ventricular interdependence.329
Because diastolic pressures are slightly lower in the right heart than the left, the RA and RV are usually the first chambers to exhibit evidence of increased intrapericardial pressure. High intrapericardial pressure can cause compression or collapse of right heart chambers.329,330 Invagination of the right atrial wall during atrial systole is a sensitive (but not specific) sign of tamponade (Fig. 15–138).330 Diastolic collapse or "buckling" of the RV free wall is a more specific sign of tamponade, and can be visualized both on 2D and M-mode imaging329 (Fig. 15–135B and C). In cases of localized tamponade or severe RVH, left atrial or ventricular diastolic collapse may be the first sign of tamponade.331
Doppler echocardiographic recordings in patients with tamponade have demonstrated an enhancement or exaggeration of the normal respiratory variation in ventricular inflow and outflow. Thus, transmitral and LVOT velocities decrease significantly with inspiration, most likely because of enhanced ventricular interdependence and a marked decrease in the transmitral diastolic gradient during inspiration (Fig. 15–139). The latter is caused both by high intrapericardial pressure as well as leftward motion of the interventricular septum from increased RV filling. Although cardiac tamponade remains a clinical diagnosis, echocardiography has significantly improved the detection of hemodynamic effects from pericardial fluid, especially in early and equivocal cases. Studies have also indicated that when echocardiography is used to direct pericardiocentesis to the site of greatest fluid accumulation, the risks associated with blind pericardial puncture are decreased.
The diagnosis of constrictive pericarditis is sometimes difficult to establish, even by cardiac catheterization. 2D and M-mode echocardiography may provide evidence of thickened pericardial tissue by demonstrating increased reflectivity and multiple parallel moving echoes in the area of the pericardium. The criteria for pericardial thickening on echocardiogram are imperfect, however, as the normal pericardium is an echodense, highly reflective structure with a gain-dependent signal.332 Paradoxical septal motion may be seen on M-mode with constriction, as can an abnormal inspiratory interventricular septal "bounce"333 and limited diastolic motion of the posterior LV wall. A dilated inferior vena cava that does not collapse on deep inspiration is indicative of high RA pressure and may be observed on 2D imaging in constrictive pericarditis.333
The utility of Doppler recordings in evaluating constrictive pericarditis has been shown in several studies.265,325,334 As with cardiac tamponade, pericardial constriction produces exaggerated respiratory variation in the isovolumic relaxation time and in flow velocities within right and left ventricles, pulmonary veins, and hepatic vein.325 A respiratory variation of >20 percent in peak mitral E velocity favors the diagnosis of constriction over restrictive cardiomyopathy, while little respiratory variation favors restrictive physiology.325 Doppler echocardiographic criteria for constriction have been validated prospectively and may help predict clinical response to pericardiectomy. Unfortunately, exaggerated respiratory flow variation is not specific for pericardial constriction and also can be seen in chronic obstructive pulmonary disease and asthma. In these cases, Doppler examination of superior vena cava flow is useful: patients with asthma will have increased flow toward the heart during inspiration, while limited forward flow will be seen in constriction (the echocardiographic equivalent of Kussmaul's sign).335 In addition, tissue Doppler imaging of the mitral annulus is useful in differentiating constrictive vs restrictive physiology. As discussed in the above section on Doppler assessment of diastolic function, both of these forms of diastolic dysfunction typically exhibit restrictive patterns on transmitral (E > A) and pulmonary venous (S < D) spectral flow recordings. In cases of pericardial constriction, however, the early diastolic velocity of the mitral annulus (Em) remains normal. In restrictive cardiomyopathy, this velocity is abnormally low. Finally, respiratory variation in the peak velocity and duration of continuous-wave Doppler TR spectral envelopes appear to reflect accurately the enhanced ventricular interdependence seen in constrictive pericarditis.336