Constrictive Pericarditis Workup

• No laboratory data are diagnostic of constrictive pericarditis. However, as a result of the universal findings of a chronically elevated right atrial pressure and passive congestion of the liver, kidneys, and gastrointestinal tract, resultant abnormalities may be present. These include elevations in both conjugated and unconjugated bilirubin levels and elevated levels of hepatocellular function tests.
• The level of brain natriuretic peptide (BNP), a cardiac hormone released in response to increased ventricular wall stretch, is often mildly increased in constrictive pericarditis (usually < 150 ng/L). BNP levels are generally higher in restrictive cardiomyopathy (diagnostic if >650 ng/L) and may be useful in differentiating these disorders.^{[11, 12]}
• Hypoalbuminemia may be a result of both a protein-losing enteropathy and proteinuria that may approach nephrotic range.
• If active or chronic inflammation is present, nonspecific markers, such as an elevated sedimentation rate or a normocytic normochromic anemia, may be present.
• If an associated collagen vascular disorder is suggested, antinuclear antibody or rheumatoid factor should be measured.
• Results from a purified protein derivative skin test should be positive in cases of tuberculous pericarditis (unless the patient is anergic).
• Expect an elevation in the white blood cell count with an associated left shift in cases of bacterial pericarditis.
• Cytologically examining the pericardial fluid, if present, helps diagnose a malignant cause (if not otherwise apparent).

• Chest radiography

  • Although findings are insensitive for the presence of constrictive pericarditis, some classic findings are suggestive of the diagnosis when present within a compatible clinical context.
  • Severe pericardial calcification is found in 20-30% of patients (see below); however, it is not specific and does not prove pericardial constriction.^[13] Constrictive pericarditis. Anteroposterior and lateral chest radiograph from a patient with tuberculous constrictive pericarditis (arrows denote marked pericardial calcification).
  • If no significant pericardial effusion is present, the cardiac silhouette may appear normal.
  • The superior vena cava, azygous veins, or both may be dilated.
  • Pleural effusions are common and are usually bilateral.
  • Pulmonary edema is rare and might suggest other cardiac or lung diseaseConstrictive pericarditis. Anteroposterior and lateral chest radiograph from a patient with tuberculous constrictive pericarditis (arrows denote marked pericardial calcification).
• Echocardiography

  • Echocardiography has been used for many years to help diagnose constrictive pericarditis and, in particular, to differentiate it from restrictive and other cardiomyopathies. Unfortunately, no echocardiographic finding is pathognomonic for constriction. However, when all the echocardiography data are taken together within a clinical context, the likelihood of constriction can usually be accurately assessed.
  • As a general principle, pericardial imaging by echocardiography is not sensitive and is not considered a reliable technique to visualize the pericardium. Admittedly, the pericardium can be echodense, but this is not always the case. CT scanning and MRI are considered the procedures of choice for imaging the pericardium.
  • Transesophageal echocardiography is more reliable than transthoracic echocardiography for helping to detect a thickened pericardium, especially if it is thick or very echogenic, but this is not nearly as accurate as CT or MRI.
  • The posterior motion of the otherwise normal interventricular septum relative to the less compliant ventricular walls (which are encased by the pericardium) correlates with the auscultatory pericardial knock (can be seen on M-mode and 2-dimensional echocardiography as an early diastolic septal notch or "septal bounce").^[14]
  • Two-dimensional echocardiography can show evidence of right-sided pressure overload such as atrial septal shifting to the left with inspiration or dilated inferior and superior vena cavae and hepatic veins. These are nonspecific signs that can also occur in right heart failure as a result of other causes.
  • Doppler echocardiography provides important hemodynamic information.
    • Early rapid diastolic filling can be determined by interrogating forward flow at the mitral and tricuspid valve levels. These are termed the E (for early filling) and the A (for atrial filling) waves. The transtricuspid velocities show an opposite pattern to the transmitral (ie, across the tricuspid valve velocities increase with inspiration and decrease with expiration, across the mitral velocities decrease with inspiration and increase with expiration). The shortened deceleration time from these peak velocities is felt to correspond to the dip-and-plateau hemodynamics seen with limited early diastolic flow. The pulmonary vein Doppler inflow pattern also has respiratory variation, with its diastolic inflow being greater than systolic inflow, which itself may even reverse. These Doppler findings are sensitive when present, but, like many echocardiographic signs, their absence does not exclude constrictive hemodynamics.^{[15, 16]}
    • Although technically challenging, Doppler ventricular inflow patterns can help distinguish constrictive from restrictive cardiac physiology. From a Doppler perspective, constriction limits ventricular filling and enhances ventricular interaction. Conversely, restriction generally limits ventricular distensibility. Respiratory variation is usually greater in constriction than in restriction (probably because of the normal intraventricular septum), usually with over 25% changes. With restriction, often the E/A ratio is more than 2, the deceleration time is less than 150 ms, and the relaxation time is less than 60 ms. Unfortunately, when such Doppler findings are not present, the diagnostic reliability decreases. If a concomitant pericardial effusion is present, it may account for some respiratory variation. ^[13]
    • Tissue Doppler echocardiography, measures the actual endocardial and epicardial tissue velocities. Since myocardial relaxation itself is preserved in pure constrictive pericarditis, the early relaxation myocardial velocity (Ea, also known as Em) is normal, whereas it is abnormal with restriction (when intrinsic myocardial disease is present). For example, given that a normal Ea is more than 10 cm/s, a near-normal (>8 cm/s) Ea supports constriction, whereas a much lower Ea supports restriction.^[17] The newer method of speckle tracking of B-mode images measures cardiac longitudinal and circumferential deformation. Patients with constrictive pericarditis were found to have constrained circumferential deformation rather than the longitudinal constraint found in patients with restrictive cardiomyopathy.^[18]
    • Doppler interrogation can be limited if patients cannot adequately vary their respiration or if concomitant myocardial disease, atrial fibrillation, or severe lung disease (eg, chronic obstructive pulmonary disease [COPD], which can lead to false-positive findings) is present. Since Doppler transmitral inflow respiratory variation can occur in COPD, other differences must then be examined. For example, the marked increase in inspiratory superior vena cava systolic flow seen in COPD is not seen in constriction.^[19]
    • Measurements of diastolic function are load-dependent (ie, dependent on preload and afterload). If atrial and ventricular filling pressures are low, Doppler interrogation findings may be falsely negative. Likewise, if atrial and ventricular filling pressures are high, respiratory variation may be masked. In such cases, preload may be reduced with either medication or dynamic maneuvers, such as tilting the patient's head up or having the patient sit. These maneuvers may unmask respiratory variation.
• Computed tomography

  • Conventional CT scanners may not help adequately visualize the parietal pericardium. However, the parietal pericardium can be visualized well using high-resolution CT. The pericardial thickness, degree of calcification, and distribution of these findings are easily measured.
  • The normal pericardium is 1-2 mm thick. An abnormal pericardial thickness is considered 3-4 mm thick or thicker.
  • Pericardial thickening of more than 4 mm assists in differentiating constrictive disease from restrictive cardiomyopathy, and a thickening of more than 6 mm adds even more specificity for constriction.
  • Supportive findings suggesting impaired right ventricular filling include dilation of the vena cava, hepatic vein, and right atrium as well as ascites or hepatosplenomegaly.
  • False-negative results may occur if a long-standing thin pericardial scar without appreciable thickening is present. That is, normal pericardial thickness does not exclude pericardial constriction and the clinical situation must always be taken in account. Therefore, if the hemodynamics and presentation are otherwise compatible, the diagnosis must still be entertained despite unremarkable pericardial imaging.
• Magnetic resonance imaging^{[20, 21]}

  • The development of real-time, high-resolution MRI and the ability to acquire images in 50 ms or less makes MRI a sensitive method for imaging the pericardium.
  • As can be done with CT scanning, an MRI can be used to measure the pericardium for thickness, calcification, and distribution of abnormalities.
  • A thickened pericardium does not prove that constrictive pericarditis is present; it must be clinically correlated. Likewise, constriction can occur in a scarred fibrous pericardium of normal thickness.
  • Gated MRI has an advantage in determining whether pericardial fluid is hemorrhagic.
  • Obtaining CT scan images may be advantageous compared to performing MRI when pericardial calcium is particularly prominent.

• Electrocardiogram

  • Chronic pericarditis is not associated with the classic ECG findings seen with acute pericarditis.
  • Findings of acute pericarditis (see Pericarditis, Acute) generally include diffuse concave ST-segment elevation that must be distinguished from other causes of ST elevation with PR depression. In most instances of acute pericarditis, the magnitude of the ST elevation is greater than one fourth of the T-wave height in the lateral V leads. In addition, if a history of these findings exists, the later development of constrictive pericarditis should be considered.
  • Over time, even if chronic pericarditis develops, no specific ECG patterns develop. Inverted T waves may persist, or all ECG findings may resolve to normal.
  • In long-standing cases, atrial fibrillation may occur, but this is certainly nonspecific.
  • If a pericardial effusion develops, a low QRS voltage may be present in the limb and chest leads. This must be distinguished from other causes of low voltage such as long-standing myocardial infarction, pleural effusion, postoperative state, or various cardiomyopathies.
  • When electrical alternans (a beat-to-beat cyclic shift in the QRS axis that may also involve the P and T waves) is present, cardiac tamponade (see Cardiac Tamponade) must be considered.

Sometimes, despite the history, physical, lab results, and noninvasive testing, an accurate diagnosis of constrictive pericarditis cannot be made. When this diagnosis is not absolute, despite all of the available information, invasive procedures, particularly a right heart catheterization and/or endomyocardial biopsy, can help make or exclude the diagnosis.

Right heart catheterization

Traditional hemodynamic criteria for constrictive pericarditis are as follows:

1. Elevated left and right ventricular diastolic pressures equalized within 5 mm Hg
2. Right ventricular systolic pressure less than 55 mm Hg
3. Mean right arterial pressure greater than 15 mm Hg
4. Right ventricular end-diastolic pressure greater than one third of the right ventricular systolic pressure (narrow pulse pressure)

In the absence of these criteria, a diagnosis of restrictive pericarditis is favored.^[22]

In addition, as mentioned earlier, one would also expect to see an exaggerated x descent with a steep y descent on right arterial pressure waveforms (W sign) as well as the square root sign (also known as the dip-and-plateau) on right or left ventricular tracings, which distinguishes this diagnosis from cardiac tamponade. Another hemodynamic parameter to look for is Kussmaul sign, which is failure of the right arterial pressure to decrease with inspiration, but this can also be seen in right heart failure, severe tricuspid regurgitation, and systemic venous congestion.^[7]

Since the respiratory variations in intrathoracic pressures are not transmitted to the cardiac chambers in constriction, this leads to decreased left-sided filling on inspiration compared with the right side. Talreja et al took advantage of this approach in a recent study that looked at the area under the right and left ventricular curves during inspiration and expiration. They found a 100% positive predictive accuracy, 97% sensitivity, and 100% specificity for constriction with a systolic area index greater than 1.1 (comparing expiration vs peak inspiration). This was not a randomized controlled clinical study, and there was selection bias; however, this may prove to be another standard diagnostic criterion in the future.^[23]

Although these signs are useful, in practice, uncertainty always exists when attempting to diagnose constrictive pericarditis. Fluid-filled catheters render notoriously poor fidelity tracings, which can lead to a misinterpretation of the hemodynamic data. Irregular rhythms, such as atrial fibrillation, may alter ventricular filling pressures based on the varying RR intervals. Variations in respiration patterns may affect hemodynamics, and patients should be instructed to breathe smoothly and uniformly during hemodynamic recordings. The patient's diastolic filling pressures can affect hemodynamic measurements, and some authors advocate infusing isotonic sodium chloride solution if the patient's left ventricular end-diastolic pressure is less than 15 mmHg to unmask occult constrictive pericarditis. Conversely, if the filling pressures are too high, subtle respiratory variations in pressure may be missed.^[8]

Important etiologies of diastolic pressure equalization to include in the differential diagnosis are restrictive cardiomyopathy, cardiac tamponade, COPD and pneumothorax (pulmonary hyperinflation), dilated cardiomyopathy (if severe, all filling pressures may be high), atrial septal defect, and volume depletion (when all filling pressures are low).

Pericardial and endomyocardial biopsy

Occasionally, direct inspection and pericardial biopsy may be required to diagnose constriction.

If constriction is strongly suggested clinically, despite the pericardium being thin on imaging, after careful consideration, direct surgical inspection, biopsy, and pericardectomy may be required to diagnose definitively or to exclude the diagnosis.^[24]

Despite the best attempts at diagnosing constrictive pericarditis, confirming the diagnosis may be impossible until surgery. Patients and their families need to be aware of this fact and that, in some cases, surgery may be considered exploratory.

See Pericardial and endomyocardial biopsy.