|| Checking for direct PDF access through Ovid
Purpose: The new 3D speckle tracking echocardiography generates simultaneous volumetric and mechanical left ventricular (LV) data because of its unique capability of quantifying dynamic 3D data sets. However, besides validation of any novel software for routine clinical application, robustness of the method needs to be investigated, the latter being the rationale of the study.Methods: One hundred and twenty six subjects were included in the study (Mean age 58 ±14 years, 53 M). Real time 3D echo images obtained on a GE system were post processed using the 3D STE software (Tomtec). Of these 20 subjects (10 normal and 10 abnormal) were analysed in duplicate by 2 blinded observers. LV end-systolic, end-diastolic, and stroke volumes (EDV, ESV, SV), LV ejection fraction (LV EF%), global LV longitudinal strain (GLS%), systolic dyssynchrony index (SDI), LV torsion (degree), and LV twist (degree/cm) were computed and the data were tested for reproducibility using inter rater agreement test for weighted kappa value as a measure of robustness of the data obtained by the observers between and within themselves 30 days apart.Results: For the inter-observer measures best agreement was seen for LVEF (Kappa 0.73, SE 0.08, 95% CI, 0.6-0.9) while the least agreement was noted for LV twist (0.24). For the intra-observer data ESV yielded the greatest agreement (0.78) with similarly poor agreement for LV twist (Fig). Global longitudinal S% had near identical agreement (0.67 and 0.69).Conclusion: 3DSTE provides fairly reproducible volumetric and strain data. Whether poor agreement for left ventricular rotational function is because of limited frame rates of 3D data set acquisition is a matter that requires further investigation.