Because of its unique hemodynamic features, constrictive pericarditis may be identified using 2-dimensional and Doppler echocardiography. An important component of a “constriction-protocol” echocardiogram is the tissue Doppler assessment of the early diastolic mitral annular relaxation velocity (e′), which provides a noninvasive measure of left ventricular myocardial relaxation. The e′ velocity is reduced in most forms of heart failure related to myocardial disease, and the ratio of transmitral velocity (E) to e′ allows estimation of left ventricular filling pressure in many cases. Specifically, an E/e′ of <8 correlates with normal filling pressure, an E/e′ >15 correlates with increased filling pressure, and an E/e′ of 8 to 15 is indeterminate with a wide range of possible filling pressures.

In contrast, the e′ velocity and E/e′ ratio in constrictive pericarditis are opposite to what is expected. Rather than being reduced, the e′ is preserved or even increased in constrictive pericarditis. Accordingly, the E/e′ ratio does not have the expected relation with left ventricular filling pressure. We coined the term “annulus paradoxus” after observing an inverse relation between E/e′ and pulmonary capillary wedge pressure (PCWP), such that the lower E/e′ ratios correlated with higher filling pressure and vice versa.

We recently evaluated e′, E/e′, and other echocardiographic variables in 130 patients with confirmed constrictive pericarditis (of mixed etiologies) compared with 36 patients with confirmed restrictive cardiomyopathy or severe tricuspid regurgitation. Differences between the groups were pronounced: the constriction group had an average medial e′ of 12.9 cm/s and an average E/e′ of 5.8, whereas the nonconstriction group had an average medial e′ of 7 cm/s and an average E/e′ of 16.1 (p <0.001 for both variables). Medial e′ velocity was found to be independently associated with a diagnosis of constrictive pericarditis on multivariate analysis. In light of this strong evidence, we have found the paradoxical characteristics of the mitral annulus to be invaluable in diagnosing constrictive pericarditis and differentiating it from restrictive cardiomyopathy. If a patient has symptoms and signs of heart failure in the setting of a preserved ejection fraction, constrictive pericarditis is the most likely diagnosis if the medial e′ velocity is preserved or increased.

We therefore read with interest the study by Alrais et al exploring the relation between echocardiographically estimated and invasively measured filling pressures in constrictive pericarditis. The investigators found no significant correlation between mean, medial, or lateral E/e′ and nonsimultaneous PCWP in a retrospective study involving 49 patients with surgically confirmed constrictive pericarditis. They concluded that E/e′ “cannot be relied upon as an estimate of filling pressures” and “should not be considered.”

We agree that we need to be cautious in estimating LV filling pressure with E/e′ in patients with constrictive pericarditis, as the relation is likely to vary with many factors including age, etiology, pericardial thickness, and superimposed myocardial disease. However, we have concerns about dismissing E/e′ from the diagnostic workup, as appears to be suggested.

The first concern is that the reported data may not accurately address the question of whether E/e′ and PCWP are inversely correlated in constrictive pericarditis. The recordings were not simultaneous, and some of the patients appeared to have normal or near-normal PCWP, which would be unusual in heart failure due to constrictive pericarditis and suggests that the patients may have been hypovolemic at the time of catheterization. Although diuretic dosing may not have changed, as specified by the authors, this would not rule out the possibility that the patients’ volume status had changed by the time of catheterization in the fasting state. Simultaneous measurement of PCWP and E/e′ would be required to definitively determine whether there is an inverse correlation.

The second concern is that failure to find a statistically significant correlation between E/e′ and nonsimultaneous PCWP does not negate the clinical utility of the E/e′ ratio. Most of the patients had low E/e′ ratios, which is exactly the “paradoxical” finding that is expected in constrictive pericarditis, compared with restrictive cardiomyopathy and other causes of heart failure. In other words, both e′ and E/e′ now have “stand-alone,” well-substantiated roles in the diagnosis of constriction regardless of correlation with invasively measured pressures.