Speckle-tracking echocardiography for diagnosis of cardiac sarcoidosis: Correlation with CMR
23 February 2017 (online)
Given the heterogeneity of sarcoidosis per se and cardiac sarcoidosis (CS) in particular, major efforts are undertaken to identify imaging modalities that allow for a reliable and timely CS diagnosis. With the intent to define its validity for the diagnosis of cardiac sarcoidal affection, we performed offline speckle tracking analysis of both CS-positive and CS-negative patients.
Study population comprised 60 patients (47.8 ± 11.4yrs, 51.7% male) with histologically proven sarcoidosis of whom 19 patients offered cardiac sarcoid involvement, whereas the remaining 41 patients absented cardiac affection. CS-presence and absence were defined on the basis of prior cardiovascular magnetic resonance (CMR) imaging. At the time of CMR-conduction, patients underwent complementary 2D transthoracic echocardiography.
Evaluation of global and regional biventricular deformation properties derived from 2D longitudinal strain imaging revealed impairment of longitudinal performance in CS-positive patients as compared to CS-negative individuals. Concretely, the left ventricular global, medial septal and apical septal longitudinal strains were significantly reduced in and correlate with CS-presence (p = 0.009, p = 0.007, p = 0.02, respectively). Comparative ROC analysis identified the left ventricular global longitudinal strain to have the highest predicted value for CS presence: the area under the curve accounted for 0.72 (95% CI: 0.56 – 0.87; p < 0.01) and was indicative of an overall fair accuracy. A cut-off value of -14.1% showed a sensitvity of 73.7% and a specificity of 73.2%. In contrast to the left ventricular strains, none of the right ventricular deformation values evidenced correlation with CS-status.
Speckle tracking analysis identified cardiac affection to correlate with reduction in left ventricular longitudinal deformation capabilities and might be an appropriate tool for timely CS-diagnosing and disease's monitoring.