학술논문
MODERATED POSTER SESSION: New imaging techniques in classical scenarios: Saturday 6 December 2014, 08:30-12:30 * Location: Moderated Poster area
Document Type
Article
Author
Fukuyama, A; Takeuchi, M; Wu, V; Nagata, Y; Hayashi, A; Otani, K; Fukuda, S; Yoshitani, H; Lin, F; Otsuji, Y; Islas, F; Olmos, C; Almeria, C; Rodrigo, JL; De Agustin, A; Marcos-Alberca, P; Clavero, M; Saltijeral, A; Perez De Isla, L; Atout, W; Centre, Cairo University Hospitals- Echocardiography Unit Diagnostic U/S; Maceira Gonzalez, A M; Igual, B; Cosin-Sales, J; Diago, JL; Aguilar, J; Ruvira, J; Sotillo, J; Bertella, E; Baggiano, A; Loguercio, M; Mushtaq, S; Petulla', M; Segurini, C; Conte, E; Andreini, D; Pontone, G; Tong, L; Ramalli, A; Tortoli, P; Luo, J; D'hooge, J; Galanti, G; Toncelli, L; Stefani, L; Pedri, S; Pedrizzetti, G; Kaminska-Kegel, AB; Jaroch, J; Brzezinska, BE; Kruszynska, E; Kusmierz, M; Loboz-Grudzien, K; Hagendorff, A; Stoebe, S; Tarr, A; Pfeiffer, D; Fazlinezhad, A; Fazlinezhad, AFSOON; Azimi, SARA; Vejdan Parast, MOHAMM; Hashemi Doost, AFSANE
Source
European Journal of Echocardiography; December 2014, Vol. 15 Issue: Supplement 2 pii232-ii232, 1p
Subject
Language
ISSN
15252167; 15322114
Abstract
Purpose: Although 2D strain is widely used to assess left ventricular mechanics, strain values derived from vendor specific 2D strain software are different even in the same subjects and are not interchangeable. We hypothesized that vendor-independent software will provide lower inter-vendor variability of 2D strain measurements. Methods: We acquired two sets of three apical images using 2 of 3 ultrasound machine (Philips, GE and Toshiba) in 81 healthy volunteers (GE vs. Philips in 26, Philips vs. Toshiba in 31 and GE vs. Toshiba in 24). Regional 2D longitudinal strain was measured using vendor-specific software and 2 vendor-independent software (TomTec and Epsilon) in each set of apical images. Then, global longitudinal strain (GLS) values were directly compared between vendor specific software and vendor independent software. Results: Correlation of GLS values between vendor specific software showed wide range of r values (0.23, 0.42 and 0.72) with significant bias except one comparison. Vendor independent software provided modest degree of correlation (TomTec: 0.65, 0.66 and 0.77, Epsilon: 0.64, 0.74 and 0.77) without significant bias in one and with significant bias in the other. Conclusions: Although vendor independent 2D strain software provided moderate correlation of GLS values between ultrasound images obtained from different ultrasound machine in the same subjects, relatively large limit agreement had a still major limitation. Our results suggest that the same ultrasound machine and the same 2D speckle tracking software should be used for the longitudinal analysis of strain values in the same subjects as well as cross sectional study. Comparison (Machine- Software) r bias LOA p (bias) Philips-Philips vs. GE-GE 0.72 -0.80 -1.81 to 3.41 0.0060 Philips-TomTec vs. GE-TomTec 0.77 0.70 -3.06 to 4.46 0.0806 Philips-Epsilon vs. GE-Epsilon 0.74 -1.17 -1.89 to 4.23 0.0010 Philips-Philips vs. Toshiba-Toshiba 0.42 -0.39 -4.17 to 3.39 0.2823 Philips-TomTec vs. Toshiba-TomTec 0.65 0.15 -4.37 to 4.68 0.7020 Philips-Epsilon vs. Toshiba-Epsilon 0.77 0.50 -2.79 to 3.78 0.1281 GE-GE vs. Toshiba-Toshiba 0.23 1.89 -1.07 to 4.85 <0.0001 GE-TomTec vs. Toshiba-TomTec 0.66 0.65 -2.46 to 3.77 0.1356 GE-Epsilon vs. Toshiba-Epsilon 0.64 1.26 -1.80 to 4.29 0.0032