|| Checking for direct PDF access through Ovid
Purpose: Left ventricular (LV) volumes and strain obtained by 3D speckle tracking echocardiography requires investigation to assess the relative discriminatory power of one over the other that has not been studied in a clinical laboratory, thereby justifying the rationale of this study.Methods: 38 human subjects with normal echocardiogram and 88 subjects (58 ±14 years, 53 M) with abnormal echocardiogram (structural and/or coronary artery disease) underwent 3 D speckle tracking echocardiography (3D-STE) using a custom SW (Tomtec). The data obtained were end- systolic and diastolic volumes (ESV, EDV), LVEF, 3D strain, 3D circumferential strain (CS %) and 3D radial strain (RS %). Objective of the study was to assess the comparative diagnostic power of the 3D STE data to identify diseased LV.Results: ROC analyses yielded areas under the curve (AUC) values for ESV (0.85), LVEF (0.84), 3D strain (0.81), 3D RS%(0.79), 3D CS%(0.76), and EDV (0.76). Pairwise comparison of the AUC showed that ESV had a significantly greater AUC than EDV (p<0.0001), and global CS% (p<0.05). LVEEF was however superior to GCS% (p<0.01). 3 D strain differed from global CS% (p<0.05) as well. 3D- LVEF could be predicted from linear regression equation using both 3D global strain (y = 6.8376 + -1.2022 x, R2= 0.78, P<0.0001) as well as by 3 D circumferential strain (y = 16.2762 + -1.3113 x, R2 = 0.78, P<0.0001). Conclusion: The association of LVEF with the strain indicates the validity of 3 D strain as a possible surrogate for LVEF with a fairly acceptable regression value. The data also suggest that 3D STE is a robust technique and that the volumetric parameters (ESV, LVEF) appear to be more powerful to discriminate normal from abnormal LV compared with 3D-STE strain obtained from multiple myocardial vectors.